Extracellular vesicles in tumor immunity: mechanisms and novel insights

Abstract

Extracellular vesicles (EVs), nanoscale vesicles secreted by cells, have attracted considerable attention in recent years due to their role in tumor immunomodulation. These vesicles facilitate intercellular communication by transporting proteins, nucleic acids, and other biologically active substances, and they exhibit a dual role in tumor development and immune evasion mechanisms. Specifically, EVs can assist tumor cells in evading immune surveillance and attack by impairing immune cell function or modulating immunosuppressive pathways, thereby promoting tumor progression and metastasis. Conversely, they can also transport and release immunomodulatory factors that stimulate the activation and regulation of the immune system, enhancing the body’s capacity to combat malignant diseases. This dual functionality of EVs presents promising avenues and targets for tumor immunotherapy. By examining the biological characteristics of EVs and their influence on tumor immunity, novel therapeutic strategies can be developed to improve the efficacy and relevance of cancer treatment. This review delineates the complex role of EVs in tumor immunomodulation and explores their potential implications for cancer therapeutic approaches, aiming to establish a theoretical foundation and provide practical insights for the advancement of future EVs-based cancer immunotherapy strategies.

Introduction

Cancer has emerged as a critical risk factor impacting human life, health, and well-being, particularly in the context of rapid social and economic development. According to the latest Global Cancer Report 2020 published by the International Agency for Research on Cancer (IARC), it is projected that there will be nearly 19.3 million new cancer cases and almost 10 million cancer-related deaths worldwide in 2020 [1]. Cancer is a complex systemic disease characterized by tumor growth that is influenced by both non-specific and specific immune mechanisms [2–4]. The processes involved in cancer development and metabolism induce various alterations in the immune system, enabling tumors to evade immune surveillance and, in some cases, to manipulate the immune response to promote their own growth [5–7]. Therefore, understanding the ongoing interactions between tumor cells and the immune microenvironment is essential for a comprehensive understanding of tumor progression.

Extracellular vesicles (EVs) are nanoparticles released by living cells containing a variety of molecules such as proteins, lipids, mRNAs, miRNAs, and DNA fragments that reflect the characteristics of the originating cell [8, 9]. These extracellular vesicular contents play critical roles in various pathological processes, such as tumorigenesis, growth, invasion, metastasis [10–17], aberrant neovascularization [17–21], epithelial-mesenchymal transition [22], immune tolerance, immune escape [23], and therapeutic resistance [10]. According to the MISEV2023 guidelines, “extracellular vesicles” are defined as particles that are naturally released from cells. These vesicles are characterized by their bilayer structures composed of lipids and include exosomes, ectosomes, exomeres, and microvesicles. In the absence of specific differentiation among subcellular sources, it is currently recommended to refer to these entities collectively as extracellular vesicles [24].

EVs, as multifunctional informational complexes, play a crucial role in mediating communication between innate and adaptive immunity, modifying receptor cell genes, reshaping the tumor immune microenvironment, and exerting either immune-enhancing or inhibitory effects to regulate cancer progression and metastasis. They have the ability to establish an immunosuppressive and pro-tumorigenic microenvironment that facilitates cancer growth [10, 23]. Tumor-derived exosomes (TEXs) can promote tumor immune evasion by disrupting the function of immune cells such as T cells, NK cells, and dendritic cells. Moreover, EVs carrying the programmed cell death ligand 1 (PD-L1) protein can aid cancer cells in evading immune surveillance by deactivating immune cells. The interaction between EVs and the tumor microenvironment (TME) is bidirectional, and the secretion of EVs is influenced by external factors. Hypoxia and elevated extracellular acidification are prevalent features of solid tumors [25, 26]. Research suggests that a hypoxic environment can stimulate tumor cells to produce and release an increased quantity of EVs, resulting in modifications to the EV cargo profile that are enriched with pro-tumorigenic factors, thereby facilitating various tumor-related processes [22, 27]. Similarly, in acidic conditions, the release and uptake of EVs by tumor cells are significantly enhanced, and the composition of these EVs is also altered [26, 28].

Recent studies suggest that EVs reprogramming could be a promising strategy for targeted immunotherapy in cancer. Research indicates that removing PD-L1 from EVs can effectively inhibit tumor growth, even in PD-L1 antibody-resistant tumors [29]. On the other hand, EVs derived from genetically modified T cells expressing chimeric antigen receptor (CAR) exhibit enhanced levels of cytotoxic molecules and lack programmed cell death protein 1 (PD1), demonstrating their superior therapeutic potential [30]. This highlights the extensive research opportunities in understanding the role of EVs in tumor immunity. Moreover, EVs, with their unique double membrane structure and multiple adhesion proteins, serve as efficient drug carriers that can deliver therapeutic agents to specific cells or tissues, opening up new avenues for tumor immunotherapy. For instance, by loading galectin-9 siRNA internally and modifying oxaliplatin externally with BM-MSC-derived EVs, targeted immunotherapy for pancreatic ductal adenocarcinoma can be achieved [31].

This review delves into the connection between EVs and the tumor immune microenvironment, emphasizing their significant impact on tumorigenesis and progression, and explores the potential of EVs as clinical therapeutic options, as well as diagnostic and prognostic biomarkers for tumors.

Functions of EVs in tumor

EVs are small membrane structures released by most eukaryotic cells from their internal compartments. They are commonly found in various biological fluids, including plasma, urine, saliva, and milk. EVs can be classified into four primary types based on size and formation mechanisms: classical exosomes (30–150 nm), microvesicles (100-1,000 nm), apoptotic bodies (100-5,000 nm), and oncosomes (1–10 μm) [32, 33].

Initial hypotheses proposed that EVs function primarily as cellular waste containers, discharging excessive or nonfunctional cellular components. However, recent studies have demonstrated that EVs can transport bioactive molecules, including nucleotides, proteins, lipids, and other biomolecules from donor cells. This capability allows them to serve as mediators of both proximal and remote intercellular communication [12, 34, 35], facilitating exchanges that contribute to various disease processes, such as neurodegenerative diseases, cancer, metabolic disorders, and cardiovascular diseases [8]. EVs are now acknowledged as significant agents in intercellular communication and signaling, with implications for both physiological and pathological processes. Notably, there is growing evidence highlighting their essential role in regulating tumor immunity and influencing tumorigenesis and progression [36].

The structure of EVs

EVs were first identified in 1983 when it was observed that reticulocytes release 50 nm vesicles containing transferrin receptors into the extracellular space [37]. Subsequent research has demonstrated that EVs are heterogeneous, reflecting the originating cell type, cellular state, stage of differentiation, and environmental stimuli present during their formation. These vesicles display a variety of cell-specific antigens on their surfaces, including fusion proteins, adhesion molecules, and integrins, which play a crucial role in their targeting of specific recipient cells [38]. Notably, the repeated invagination of the lipid bilayer during the formation of EVs leads to an orientation of the lipid bilayer that mirrors that of the cell of origin [39].

EVs, akin to their origins, are composed of lipid bilayers that can transport a diverse range of metabolites, including proteins, lipids, nucleic acids, and other cellular metabolites. These components are well-documented in the ExoCarta database [40–42]. Proteomic analyses have demonstrated that EVs encompass a wide array of proteins, including those derived from endosomal, plasma, cytoplasmic, and nuclear sources. The complex protein composition of EVs is thought to be intricately linked to their biogenesis pathways, functions, and cellular origins [43]. These proteins can be broadly classified into general and specific categories. General proteins, such as CD63, CD81, CD9, tetraspanins, heat shock proteins (HSP70, HSP90), and integrins, are ubiquitously present in EVs across various cell types [33, 44, 45]. The prevalence of these proteins in EVs may stem from their higher expression levels in primitive cells, and they serve multiple roles, including intercellular signaling, immunomodulation, and cell adhesion, thereby reflecting the cellular state and function.

Many components of EVs may reflect common pathways of EV biogenesis. For instance, tetraspanins are non-specific proteins that can interact with and form complexes alongside integrins or major histocompatibility complex (MHC) molecules. Conversely, specific proteins such as the tumor-associated high expression of epithelial cell adhesion molecule (EpCAM), CD44 [46–49], epidermal growth factor receptor (EGFR) [50–52], HER2 [53], and MHC II, which is found in dendritic cells and B-lymphocyte EVs [54, 55], are noteworthy. Additionally, proteins involved in membrane transport and fusion, such as Rab GTPases and membrane-associated proteins, along with those implicated in EV biosynthesis—like the endosomal sorting complex required for transport (ESCRT), ALIX, and TSG101—are also present. The membranes of EVs may be enriched with specific membrane proteins that facilitate their formation, release, targeting, delivery, and biological activity, thereby providing insights into the cell of origin and the associated disease state [56–58]. For example, EVs derived from B-cell lymphoma cell lines display distinct patterns of surface protein expression, characterized by variations in the levels of CD19, CD20, CD24, CD37, and HLA-DR, in contrast to CD22, CD23, CD40, and CD45. Additionally, there is considerable heterogeneity in surface protein expression among different B-cell lymphoma cell lines. Importantly, these B-cell lymphoma-derived extracellular vesicles have been shown to protect lymphoma cells from complement-dependent cytotoxicity induced by rituximab [59], underscoring the significance of this finding in elucidating the biology of B-cell lymphoma and in the development of novel therapeutic strategies.

EVs exhibit a distinct protein expression profile, notably including Tau proteins that are associated with Alzheimer’s disease. Research has demonstrated that EVs play a significant role in the transport of Tau proteins, which are essential to the progression of Alzheimer’s disease [60]. Furthermore, studies involving nasopharyngeal carcinoma cell lines, clinical specimens, and in vivo experiments using a zebrafish model have confirmed the presence of the nuclear protein High Mobility Group Box 3 (HMGB3) within EVs. This finding underscores the involvement of EVs in tumor angiogenesis during the metastasis of nasopharyngeal carcinoma [61].

The initial interest in EVs was ignited by their role in the transport of extracellular RNA (exRNA) [16, 62]. Researchers have identified a variety of RNA species within EVs, including messenger RNAs (mRNAs), microRNAs (miRNAs), transfer RNAs (tRNAs), long noncoding RNAs (LncRNAs), and viral RNAs. These RNA species have the potential to facilitate the transfer of genetic information between cells via EVs, thereby influencing gene expression and the functional characteristics of recipient cells [14, 16, 19–21, 62–65]. This intercellular communication can significantly impact cellular processes such as migration, apoptosis, proliferation, and autophagy. Furthermore, EVs hold considerable promise in the realms of disease diagnosis, efficacy prediction, and prognosis assessment [66–70]. In vitro translation experiments have demonstrated that extracellular vesicular RNAs can traverse between cells and modulate target mRNAs at distinct locations, leading to the concept of ‘exosomal shuttle RNAs’ (esRNAs) [62, 71]. Although the precise mechanisms governing RNA sorting into EVs remain incompletely understood, existing evidence indicates that this sorting process is not random. For example, research conducted by Marco Tripodi has highlighted the RNA-binding protein SYNCRIP as a crucial factor in the sorting of miRNAs within hepatocyte-derived EVs; a reduction in SYNCRIP levels adversely affects miRNA sorting. Additionally, another protein, heterogeneous nuclear ribonucleoprotein A2B1 (hnRNPA2B1), appears to play a similar role in this sorting process [72].

Utilizing the Hypergeometric Optimization of Motif Enrichment (HOMER) for RNA sequencing, Kahn’s research elucidates that microRNAs (miRNAs) contain specific sequences, referred to as EXOmotifs and CELLmotifs, which govern their secretion or retention within cells. It is evident that altering these sequences in miRNAs can significantly affect their release from EVs, as well as their intracellular retention. Notably, miRNAs that include the CGGGAG motif, classified as an EXO motif, are regulated by RNA-binding proteins Alyref and Fus, facilitating the delivery of miRNAs and the subsequent suppression of target genes in distant cells [73].

Numerous research teams have identified that the nuclear exosome targeting (NEXT) complex, which comprises ZCCHC8, the RNA deconjugating enzyme MTR4, and the RNA-binding protein RBM7, plays a critical role in regulating RNA entry into EVs [74, 75]. This complex functions synergistically to bind and translocate RNA into EVs, thereby facilitating the degradation of short, non-sequence-specific nuclear RNAs [76]. Furthermore, the activity of the NEXT complex is modulated by various cancer-associated proteins, including MYCN-recruited EVs, which enhance transcription elongation during the S phase. This process is crucial for preventing transcription-replication conflicts and supports the rapid proliferation of neuroendocrine tumor cells [77]. Collectively, these findings elucidate the mechanisms underlying miRNA sorting into exosomes, providing valuable insights into the relationship between circulating exosomal miRNAs and their tissue origins, and suggesting innovative strategies for optimizing RNA-based therapeutic interventions.

EVs have been demonstrated to transport DNA, including intact mitochondrial genomes, in patients with hormone therapy-resistant metastatic breast cancer [78]. The debate surrounding the presence of cytosolic DNA in EVs arises from their origin in cytoplasmic multivesicular bodies (MVBs), rather than the nucleus. Nevertheless, studies have identified substantial fragments of double-stranded DNA (> 10 kb) encompassing all chromosomes within EVs [79], a finding corroborated by multiple investigations [80–82]. Furthermore, nuclear extracellular vesicles (nEVs), which are enriched with nuclear proteins and genomic DNA, have been identified [83], with cancer cells secreting nEVs in greater quantities compared to healthy cells [84]. Despite this, EVs are not regarded as active carriers of DNA release [33]. Recent research suggests that EV secretion is vital for maintaining cellular homeostasis; inhibiting this secretion leads to the accumulation of cytoplasmic DNA and triggers various cellular responses, ultimately resulting in cell cycle arrest or apoptosis [85]. The presence of chromosomal DNA fragments in EVs implies that EV secretion serves to eliminate deleterious cytoplasmic DNA, thereby providing new insights into the biology of EVs and their role in the regulation of cellular homeostasis [85].

The membranes of EVs are predominantly composed of various lipids, including cholesterol, sphingomyelin, glycosphingolipids, phosphatidylcholine, phosphatidylethanolamine, phosphatidylinositol, phosphatidylserine, and saturated fatty acids [86, 87]. Research indicates that while EVs derived from different cell lines exhibit similar lipid class compositions [88], the relative proportions of these lipids can vary based on specific conditions and sources [89]. These lipids play a crucial role not only in the formation of EVs but also in influencing their stability, functionality, and participation in intercellular signaling processes [89]. The membranes of EVs demonstrate remarkable stability in bodily fluids and across varying pH levels, attributed to the presence of sphingolipids and phosphatidylinositol, which confer protection against degradation by proteolytic or lipolytic enzymes. Furthermore, EVs harbor a diverse array of small-molecule metabolites, hormones, vitamins, fibronectin, hyaluronic acid, and polysaccharides, all of which contribute to the regulation of intercellular communication, the structural integrity of the extracellular matrix, and facilitate the isolation of EVs [90].

The biogenesis of EVs

The production and release of EVs is a complex process that involves a range of intracellular and extracellular molecular mechanisms (Fig. 1). EVs can originate from various cell types, including tumor cells, immune cells, and nerve cells, each of which develops specific organelles or regions dedicated to EV generation. In contrast to the outward budding of ‘extracellular bodies’ or ‘shedding vesicles’ [91], the biogenesis of EVs begins with the inward budding of the plasma membrane, a process initiated by neurocortin, resulting in the formation of early endosomes. These early endosomes subsequently undergo further invagination, leading to the development of late endosomes and MVBs that contain numerous intraluminal vesicles (ILVs) with diameters ranging from 40 to 150 nm, which are collectively referred to as EVs. During maturation, these vesicles encapsulate various biomolecules, including proteins, RNAs, and lipids. The selection and aggregation of these molecules are likely influenced by the cellular state, environmental factors, and the specific cell type. EVs are released when late-stage MVBs fuse with the plasma membrane, a process that may be facilitated by multisubunit tethering complexes (MTCs), SNARE proteins, and Sect. 1/Munc18-like (SM) proteins [92–95]. While some MVBs are directed toward lysosomes for degradation, others may fuse with autophagosomes to form amphisomes, which can either merge with lysosomes for degradation or with the plasma membrane to facilitate EV secretion [32, 54].

Fig. 1.

Biogenesis of EVs. EVs encompass a diverse range of cargo, including siRNA, circRNA, lncRNA, mRNA, miRNA, lipids, and proteins, all of which play critical roles in various biological mechanisms within cells. With a size range of 30–150 nm, the proteins found in EVs are likely to be pivotal in the biogenesis of these vesicles. Targeting these proteins may provide a means to modulate EV biosynthesis, thereby offering valuable insights for the development of therapeutic strategies

The formation of MVBs and the subsequent outgrowth and sorting of vesicles into EVs can be classified into mechanisms that are either dependent on the ESCRT or non-ESCRT-dependent. The specific mechanisms underlying EV biogenesis are influenced by factors such as cargo, cell type, and the cellular environment. Key proteins, including the Rab family of GTPases and the SNARE complex, play critical roles in MVB transfer and fusion [92, 96]. Notably, RAB proteins [97], such as RAB27A, RAB27B, RAB11, RAB35, and RAB7, are integral to regulating EV secretion, with their functions being context-dependent. For instance, intratumoral hypoxia has been shown to upregulate the expression of the RAB22A gene, which results in increased microvesicle shedding and enhanced tumor metastasis [98]. In specific cell lines, including HeLa cells, RAB27A and RAB27B are essential for EV release [99], with RAB27A being particularly significant in melanoma [100] and 4T1 breast cancer cells [101]. Furthermore, RAB31 has been identified as having a dual role in non-ESCRT-dependent EV formation, as it promotes the formation of ILVs while simultaneously inhibiting MVB degradation [102].

SNARE proteins are essential for intracellular membrane fusion, as they facilitate the fusion of vesicles with target compartments and the delivery of cargo. These proteins are generally categorized into v-SNAREs, which are found on transporting vesicles, and t-SNAREs, which are located on the target membrane [95, 103]. The SNARE complex typically comprises three or four proteins, including synaptophysin, VAMP, and SNAP-25 homologs, with at least two of these proteins anchored to vesicles and target membranes via their carboxy-terminal transmembrane domains [104]. Research indicates that a decrease in O-GlcNAc acylation of SNAP-23 in breast cancer cell lines and mouse models enhances EV secretion, contributing to cisplatin resistance [88]. Furthermore, VPS33B has been identified as a crucial regulator of EV maturation and secretion in both hematopoietic stem cells and leukemia-initiating cells in mice and humans [105]. Additionally, the microenvironmental pH influences this process, as lower pH levels can affect integrin activity and the lipid composition of cell membranes, thereby promoting EV release and uptake by recipient cells [106, 107].

EV production is significantly dependent on the ESCRT, which identifies ubiquitinated proteins in vivo and facilitates their transport and subsequent release as EVs. Among the various pathways involved, the ESCRT-dependent pathway has been the most extensively studied [108]. The ESCRT machinery consists of four distinct complexes: ESCRT-0, which includes Hrs, Stam1, and Stam2; ESCRT-I, characterized by the presence of Vps28; ESCRT-II, comprised of Vps25, Vps36, Vps37, and Tsg101, among others; and ESCRT-III, which includes Vps2 and Alix, among other components. These multiprotein complexes, along with auxiliary proteins such as ALIX, VPS4, and VTA1, localize to the cytoplasmic side of the endosomal membrane, playing a crucial role in sorting proteins into ILVs and participating in the ubiquitination of target proteins [109]. ESCRT-0 is responsible for recognizing and retaining ubiquitinated proteins on late endosomal membranes, while ESCRT-I and ESCRT-II interact with ESCRT-0 to initiate membrane invagination into the MVB. ESCRT-III forms a helical structure that constricts the neck of the invaginated membrane, leading to membrane rupture, a process driven by the ATPase VPS4 protein [110]. This mechanism is further regulated by various cancer-associated proteins. Notably, the interaction between GPR143 and the ESCRT-0 subunit HRS not only facilitates the selective sorting of proteins into the ILVs of MVBs but also enhances EV secretion [111].

The complete depletion of all four subunits of the ESCRT complex does not entirely obstruct the formation of MVBs; rather, it modifies the structural morphology of the MVB, thereby limiting the production of ILVs that contribute to EVs [112]. In the ESCRT-independent pathway, heat shock proteins 60 (HSP60), HSP70, and HSP90 function as chaperones, while the four-transmembrane proteins CD63, CD81, CD82, CD37, and CD9 are essential for membrane formation and cargo binding. Among these proteins, CD63 and CD81 are the most prevalent in the ILV membrane and are regarded as key markers of EV release [113].

Distinct structural domains of the plasma membrane, enriched in exosomal and endosomal proteins, are present in Jurkat, SupT1, and progenitor T cells, facilitating the maintenance of endosomal characteristics essential for outward vesicle budding and the direct biogenesis of EVs. The secretion of HIV Gag, a critical factor in HIV propagation, from cellular EVs is achieved by targeting HIV Gag to the endosomal-like structural domain of the plasma membrane [114]. A similar process of spontaneous EV budding was observed in the plasma membrane substructures following the expression of the Nef protein in the aforementioned cell lines [115]. These vesicles contain characteristic extracellular vesicular markers, such as CD63 and CD81, and display comparable sucrose gradient densities and sizes in electron microscopy, making them indistinguishable from classical EVs derived from endosomes.

The presence of nEVs suggests that the mechanism underlying EV secretion may be incomplete. Following treatment with genotoxic drugs, an increase in micronuclei (MN) and nEVs was observed, indicating direct interactions with MVB precursors and EV markers, such as tetratransmembrane proteins. This observation implies a potential new mechanism for EV secretion [84]. However, this study has limitations in elucidating the biogenesis of EVs and their related mechanisms. The rate of EV biogenesis remains largely unknown and may vary based on cell type and its physiological or pathological state, an aspect that warrants further investigation. Additionally, the differentiation of MVBs into degradation pathways or fusion with the cellular membrane, as well as the regulation of the proportion of MVBs that fuse with the cell membrane via the endocytosis system, remains unclear and requires more thorough examination. The categorization of ILVs into MVBs is not well defined, and the specific mechanisms governing the transport of cellular components into EVs are not fully understood [38]. It is also uncertain whether the packaging of EVs occurs through a stochastic process or is orchestrated by specific mechanisms. The content and quantity of EVs produced may vary in response to different stressors or stimuli, and EVs derived from the same cells can exhibit diverse components. While the precise mechanisms for the selective packaging of cargo into EVs remain elusive, certain proteins consistently associated with EVs suggest a degree of specificity in protein transport. Furthermore, anchoring to the plasma membrane may play a crucial role in enriching specific proteins associated with EVs.

Mechanisms of cell-cell and intercellular communication via EVs

The efficacy of EVs in biomedical applications depends on their capacity to deliver genetic information to target cells effectively. This delivery can occur through various mechanisms, including receptor-ligand interactions, direct membrane fusion, or internalization via cytophagy. Once internalized by the target cell, EVs may either integrate into the endosome and be transported to adjacent cells through transcytosis, or they may fuse with the endosome, maturing into lysosomes for degradation [65, 116]. The bioactive molecules contained within EVs influence target cells through multiple pathways, such as direct activation via binding to surface ligands, the transfer of activated receptors to recipient cells, and epigenetic reprogramming facilitated by the delivery of proteins, lipids, and RNA. As a result, source cells can effectively communicate with specific target cells, whether they are in close proximity or at a distance, through the use of EVs.

It is widely acknowledged that nearly all cells secrete and uptake EVs. Some researchers propose that the specific functions of EVs are contingent upon their cellular origin. For example, EVs derived from tumor cells are crucial in intercellular communication, particularly in processes related to migration and invasion [111]. Furthermore, the uptake of EVs by recipient cells is not random; rather, EVs can be selectively targeted to interact with specific recipient cells via surface membrane proteins, thereby inducing distinct intracellular signaling pathways [16]. Certain studies have indicated that adhesion-related molecules present on the surfaces of EVs, such as tetraspanins, glycoproteins, and integrins, dictate which recipient cells receive the EVs [117, 118]. For instance, EVs containing tetraspanin 8 (TSPAN8) and integrin α4 are preferentially internalized by CD54⁺ pancreatic cells. The presence of TSPAN8-α4 integrin (CD49d) on EVs enhances their binding and uptake by endothelial cells, thereby promoting angiogenesis [118]. Additionally, the expression of integrin CD47 on engineered EVs facilitates their uptake by tumor cells through a mechanism known as micropinocytosis [117]. Moreover, the interaction between EVs and recipient cells results in the assimilation of ‘new’ cell membranes from the recipient cells, incorporating surface molecules from the donor cells [119]. Consequently, the targeted cells may acquire novel adhesion or cell recruitment properties. Importantly, the contents of EVs can be transferred to recipient cells through fusion, thereby influencing cellular reprogramming [120]. These findings highlight the dynamic role of EVs as mediators of intercellular communication.

Following their release from the cell, neighboring receptor cells can internalize EVs to access the information they contain. Evidence indicates that EVs are actively internalized via an endocytosis process, rather than through passive membrane fusion. This assertion is supported by several observations: reduced uptake of EVs by cells incubated at 4 °C [63, 121], the internalization of EVs surrounded by intracellular double-membrane structures [122, 123], and the co-localization of EVs with various endosomal and lysosomal markers [123]. Furthermore, this endocytosis process is intricately associated with lattice proteins, litter proteins, and lipid rafts [38, 122].

Upon contact with the cell membrane, EVs can influence the target cell through one of three mechanisms: direct interaction with plasma membrane receptors, fusion with the target cell membrane, or entry into the cell via endocytosis. The endocytosis process includes several pathways, such as phagocytosis, micropinocytosis, lipid raft-mediated endocytosis, clathrin-mediated endocytosis, and caveolin-mediated endocytosis [124, 125].

Phagocytosis is a cellular process primarily carried out by specialized cells, such as macrophages, that involves the engulfment of large particles or other cells into vesicles. During this ingestion of particulate material, the cell undergoes partial deformation, with the plasma membrane either invaginating or forming pseudopodia to enclose and internalize the particles. In contrast, receptor-mediated endocytosis is a specific uptake mechanism whereby cells internalize extracellular proteins or other compounds through surface receptors. These receptors exhibit high specificity, binding to corresponding ligands to form complexes. Following this interaction, the plasma membrane invaginates to create coated pits, which subsequently detach to form coated vesicles, thereby transporting extracellular materials into the cell. The molecular mechanisms underlying endocytosis involve various proteins and lipids, including clathrin, receptor proteins, and lipid components such as phosphatidylinositol and cholesterol. Specific ligands present on EVs can interact with membrane components to initiate endocytosis, resulting in the internalization of vesicles and their cargo [126, 127]. Distinct from the uptake of large particles via phagocytosis and receptor-mediated endocytosis, micropinocytosis entails the non-specific uptake of substances through the formation of small vesicles, typically less than 100 nm in diameter, which encapsulate fluid substances. In this process, the plasma membrane invaginates to generate small vesicles that detach from the membrane, a pathway that is particularly crucial for the uptake of EVs [128].

EVs, once internalized by recipient cells, can release their contents, thereby influencing cellular function, metabolism, and gene expression, which facilitates the transmission and regulation of biological information. Consequently, the composition of EVs is critical to their efficacy. In summary, the mechanism by which EVs transmit information encompasses the production and secretion of vesicles within cells, the uptake and release of their contents, and the subsequent response from recipient cells. This process represents a fundamental mode of intercellular communication, essential for the maintenance of tissue and organ function, and is implicated in both physiological and pathological conditions.

Through the exploration of EVs and their role in intercellular communication, researchers have identified their involvement in a diverse range of biological processes, including viral infections, immune regulation, and the development and reproduction of mammals. EVs not only serve as carriers for molecules that modulate immune system activity—such as antigens, antibodies, and regulatory T-cell-related molecules—but their molecular composition is also influenced by various disease states. This alteration results in specific molecules within EVs acting as biomarkers for disease diagnosis and monitoring [129]. Additionally, EVs have been investigated as potential vehicles for drug delivery, offering prospects for targeted delivery and enhanced efficacy through modifications to their composition [130]. As research on EVs progresses, the potential applications in biology, medicine, and biotechnology continue to broaden.

The biodistribution and role of EVs in tumorigenesis and development

The spatial and temporal heterogeneity of EVs remains ambiguous and contentious, with some studies focusing on their biodistribution. Although EVs can be produced by nearly all cell types, they exhibit significant heterogeneity, which may be further exacerbated in pathological conditions such as cancer. Conservative estimates indicate that normal human blood contains approximately 2,000 trillion EVs, whereas blood from cancer patients is estimated to contain around 4,000 trillion EVs [131].

In the tumor microenvironment, the properties of the extracellular matrix (ECM) significantly influence the secretion of EVs. Recent studies have demonstrated that a rigid ECM can markedly enhance the release of EVs from cancer cells. This process is mediated by a molecular signaling pathway in which ECM stiffness activates the Akt pathway, subsequently promoting the GTP loading of Rab8, a critical regulator of EV secretion. Notably, EVs derived from cells cultured on stiff ECM have been shown to effectively stimulate tumor growth [132]. Similarly, research indicates that in breast cancer, a rigid ECM facilitates EV secretion in a manner dependent on the YAP/TAZ pathway. Furthermore, EV-THBS1 mediates these stiffness-induced effects by involving matrix metalloproteinases and focal adhesion kinases [133]. Beyond secretion heterogeneity, studies have revealed that the uptake of EVs is influenced by surface molecular markers. In the tumor microenvironment, specific tumor cells, such as those in gastric cancer [134] and pancreatic cancer [135], release EVs that can bind to target cells through receptor-mediated mechanisms, such as CD44, thereby promoting tumor metastasis and immune evasion. Concerning EV distribution, when administered systemically, EVs primarily accumulate in tissues such as the liver, spleen, kidneys, lungs, and gastrointestinal tract. This distribution is influenced by various factors, including the cellular source of the EVs, the composition of the exosomal membrane (including proteins, lipids, and glycans), and the host’s pathophysiological condition [136]. Physiological environmental factors significantly influence the distribution and function of EVs. Prolonged stress and hypoxic conditions within the tumor microenvironment enhance the secretion of substantial quantities of EVs by tumor cells [137]. For example, in the context of cancer therapy, research has demonstrated that EVs released from irradiated esophageal squamous cell carcinoma cells induce G2/M phase arrest by modulating cyclin B1 and CDK1. Additionally, these EVs reduce apoptosis through the PI3K/AKT/FOXO3A pathway, in conjunction with Bax and Bcl2, and are involved in the repair of IR-induced DNA damage via γH2AX, thereby contributing to radioresistance [138].

EVs are heterogeneous entities that carry a variety of biologically active substances and play a critical role in cellular communication. They possess the capacity to influence the behavior of both donor and recipient cells through various communication pathways, including autocrine, endocrine, paracrine, and distal pathways [139, 140]. This influence can occur via direct cellular interactions or through the delivery of substances, ultimately affecting tumorigenesis and development in multiple ways. For instance, EVs can modulate TME, regulate angiogenesis, promote tumorigenesis and metastasis, mediate drug and radiation resistance in tumors, and enhance inflammatory responses [141, 142] (Fig. 2).

Fig. 2.

The multifaceted role of TME EVs in influencing tumor progression. Tumor development involves a range of processes, including survival, growth, angiogenesis, invasion, metastasis, drug resistance, and immune regulation. EVs are integral to these processes, significantly impacting tumor biology

In the autocrine pathway, EVs play a crucial role in regulating the growth of donor cells by transmitting information to the cell membrane or excreting cellular metabolic wastes. For example, EVs derived from chronic granulocytic leukemia cells contain the cytokine transforming growth factor (TGF) B1, which binds to the TGFβ1 receptor on leukemia cells. This interaction stimulates tumor growth through the activation of ERK, AKT, and anti-regulatory pathways in the producing cells [143]. Additionally, transmission electron microscopy using immunogold labeling has detected double-stranded DNA within certain EVs, which remain in the cell plasma until secreted [144]. The inhibition of EV secretion leads to the accumulation of nuclear DNA in the cytoplasm, resulting in cell cycle arrest or apoptosis through the activation of a reactive oxygen species-dependent DNA damage response. Thus, EVs facilitate the release of harmful cytoplasmic DNA from cells, ultimately promoting cell survival and maintaining cellular homeostasis [85].

EVs play a pivotal role in mediating intercellular interactions and regulating the tumor microenvironment. These vesicles transport cargoes that serve as external stimuli for recipient cells, resulting in modifications to their signaling pathways. The inherent heterogeneity among cancer cells leads to the activation of receptors or alterations in miRNA and RNA expression in neighboring cancer cells by EVs derived from host cancer cells, thereby influencing their biological characteristics. For example, glioma cells can transfer EVs containing the oncogenic receptor EGFRvIII to adjacent glioma cells that lack this receptor, which subsequently activates the AKT pathway in these recipient cells, facilitating their independent growth [145]. Moreover, mutant oncogenes such as KRAS, EGFR, and SRC can be transferred via EVs to wild-type KRAS receptor colon cancer cells, thereby promoting tumor invasion [146]. Additionally, breast cancer cells release EVs that contain PD-L1, which are internalized by other cancer cells exhibiting low or absent PD-L1 expression, thus aiding in tumor evasion from immune surveillance [147].

EVs are transferred not only between cancer cells but also between cancer cells and stromal cells. This transfer enables stromal cells to receive EVs from cancer cells, thereby contributing to a tumor-promoting microenvironment. Conversely, cancer cells exploit EVs released by stromal cells to enhance their proliferation and invasion. For example, TEXs stimulate the proliferation of endothelial cells and promote angiogenesis. Additionally, EVs from breast cancer cells impart specific characteristics to normal fibroblasts and epithelial-transformed cancer cells [148]. Notably, EVs derived from breast cancer cells induce the transformation of adipose-derived mesenchymal stem cells (MSCs) into tumor-associated myofibroblasts via the TGFB/SMAD signaling pathway [149]. Some cancer cells release TGFB-rich EVs that facilitate the transformation of fibroblasts into alpha-smooth muscle actin-positive myofibroblasts. Research has shown that a decrease in SIRT1 levels in senescent mesenchymal stromal cells, along with the expression of ATP-binding cassette subfamily B member 4 (ABCB4) in cancer cells, correlates with poor clinical outcomes post-chemotherapy [150]. Senescent stromal cells serve as crucial noncancerous components of tumors, exhibiting potent tumor suppressor effects. However, they can paradoxically promote cancer progression through a senescence-associated secretory phenotype (SASP) [151]. Furthermore, reduced levels of SIRT1 in senescent cells can lead to increased production of small extracellular vesicles (sEVs), which subsequently alter gene expression patterns, resulting in elevated ABCB4 levels and enhanced aggressiveness of cancer cells, particularly regarding their resistance to treatment [85]. This finding highlights a potential strategy for reducing the aggressiveness and resistance of cancer cells. It is proposed that therapeutic interventions aimed at disrupting the interactions between senescent mesenchymal cells and the surrounding microenvironment may provide novel opportunities to enhance the effectiveness of cancer therapies.

Tumor cell-derived EVs have been demonstrated to significantly influence the characteristics of endothelial cells and promote angiogenesis, particularly under hypoxic conditions. EVs containing tetra-transmembrane proteins have been identified as facilitators of tumor growth by enhancing angiogenesis. For example, EVs derived from cancer cells with elevated levels of TSPAN8 have been shown to stimulate both endothelial cell proliferation and angiogenesis by upregulating the expression of angiogenesis-related genes [152]. Additionally, cancer cell-derived EVs that carry the NOTCH ligand Delta-like 4 (DLL4) have been proven to enhance vascular density and branching in vivo [153]. In lung cancer cells subjected to hypoxia, there is an increased release of EVs enriched with miR-23a, which inhibits the targets of prolyl hydroxylase domain proteins 1 and 2 (PHD1 and PHD2), leading to the accumulation of hypoxia-inducible factor 1-alpha (HIF1A) in endothelial cells [154]. Furthermore, extracellular vesicular miR-23a targets the tight junction protein ZO-1, thereby enhancing vascular permeability and facilitating cancer cell migration. In the hypoxic bone marrow environment, EVs derived from multiple myeloma have been found to contain miR-135b, which inhibits the hypoxia-inducible factor 1 inhibitory factor (FIH1AN) in endothelial cells, ultimately promoting endothelial tube formation [155].

The aforementioned examples illustrate the tumor-promoting effects of EVs. However, it is plausible that EVs may also exhibit anti-tumor properties and contribute to the inhibition of disease progression. EVs possess the capacity to carry and transmit immune-activating signals, thereby stimulating the immune response to combat tumors. Further research into the intricate nature, diverse functions, and spatial and temporal heterogeneity of EVs is essential for understanding their potential beneficial roles in tumor biology. For instance, given their immunogenic properties, EVs could be investigated as a means of drug delivery, offering novel therapeutic strategies for cancer patients [156]. Future studies should aim to elucidate the specific mechanisms of action of EVs across different tumor types and explore technological approaches to enhance their specificity and safety. Investigating the heterogeneity of EVs is not only crucial for uncovering their complex roles in tumor progression but also pivotal for developing targeted therapeutic strategies based on EVs.

Immunomodulation plays a role in the exocrine life course

The immune system is a complex network of immune cells, tissues, and organs that work together to protect the body from diseases and infections. Additionally, it is essential for detecting and eliminating abnormal cells, including tumor cells [157, 158].

EVs play a crucial role in influencing the tumor immune response by releasing immunomodulatory molecules, such as cytokines and antibodies, that regulate the activity of immune cells in the TME. Conversely, stimuli from tumor immune cells and the microenvironment can impact the behavior of EVs, thereby influencing the tumor immune microenvironment. Factors such as the activity levels of immune cells and the stimuli they encounter can affect the quantity and composition of EVs they release. Furthermore, molecules present in the extracellular matrix and cytokines within the TME can influence both the release and chemotactic activity of EVs. The surface receptors of immune cells may express molecules that interact with extracellular vesicular surface molecules, facilitating the uptake of EVs. Additionally, the presence of tumor immune barriers within the TME may restrict the regulatory role of EVs. The following paragraphs will delve into each of these aspects of the interplay between tumor immunity and the life cycle of EVs.

Previous studies have demonstrated that cells within a TME secrete a higher quantity of EVs compared to normal cells [159, 160]. Cancer-associated cells exhibit a significantly higher secretion of EVs compared to healthy cells, driven by the necessity for intercellular communication and nutrient exchange [161, 162]. Research indicates that the concentration of EVs in the bloodstream of cancer patients is approximately double that found in healthy individuals [131]. Furthermore, tumor-derived myeloid-derived suppressor cells (G-MDSCs) release a greater quantity of EVs than their splenic counterparts [163]. This phenomenon has been observed across various cancer types, including glioma [164], breast cancer [165], hepatocellular carcinoma (HCC) [166], pancreatic cancer [167], gastric cancer [168], colorectal cancer (CRC) [169], and prostate cancer [170]. Guo and colleagues initially noted that the conversion of macrophages into tumor-associated macrophages (TAMs) is linked to a notable rise in EV release. To delve deeper into the specific mechanism behind the increased EV secretion in TAMs, the researchers first examined the activation status of Rab27a, a protein that plays a crucial role in regulating EV secretion in TAMs. Their findings revealed a significant elevation in the level of activated Rab27a (GTP-Rab27a) in TAMs compared to control cells. Subsequent investigations showed that supernatants released by tumor cells could enhance the phosphorylation of Akt protein in macrophages. Phosphorylated Akt, in turn, can activate MADD, a nucleotide exchange factor for Rab27a, consequently boosting EV secretion in TAMs by activating Rab27a [171]. Analysis of melanoma patient tissue samples demonstrated that tumor areas with high levels of phosphorylated HRS hindered CD8 + T-cell infiltration by promoting the production of immunosuppressive EVs expressing PD-L1. This phenomenon was also confirmed in a mouse tumor model, where tumors with elevated levels of phosphorylated HRS exhibited increased resistance to anti-PD-1 therapy [172].

The immune system can also influence the rate of EV secretion by directly regulating key molecules involved in the process. Activation of TLR4 by natural immune signaling triggers the release of EVs from macrophages. The release of EVs is influenced by the presence of ligands from Gram-negative bacterial cell walls, specifically lipopolysaccharide (LPS), which initially increases and then decreases during stimulation, following a clear temporal pattern [173]. This effect is mediated by the TRIF signaling pathway downstream of LPS-TLR4 activation, leading to the production of type I interferon that modulates the expression of PIP5K1C, affecting the content of PI4P on MVB and promoting EV release. Prolonged stimulation results in increased expression of HSPA5, which alters the localization of PI4P on MVB and reduces EV release. Additionally, the cGAS/STING pathway, a natural immune DNA receptor, has been found to incorporate oligomers into EVs. The STAM protein, a component of the ESCRT, interacts with STING proteins in oligomeric form and facilitates their translocation into EVs, aiding in their degradation by lysosomes [174].

Immune stimulation can modulate the content and functional properties of EVs, thereby influencing critical immune processes such as leukocyte migration, cell adhesion, and immune activation. Analysis of EVs generated from human bronchial epithelial cells stimulated with T2- and T17-type immunoreactive cytokines revealed significant differences in transcriptomic and proteomic profiles. Pathway analysis highlighted that the distinct genes and proteins were primarily linked to leukocyte migration and cell adhesion. Subsequent experiments confirmed a tendency for T17-derived EVs to promote an increase in neutrophil migration [175]. Similar observations were made in EVs from LPS-activated compared to unactivated macrophages. The enhanced components in LPS-activated macrophage EVs were predominantly associated with complement activation, regulation of reactive oxygen species, leukocyte migration and activation, and monocyte chemotaxis [176].

EVs from different cellular sources regulate tumor immunity

EVs have been demonstrated to transport immunomodulatory bioactive elements that play a role in shaping and governing the intricate network of tumor immunity. They are involved in the immune response, impacting tumor progression and acting as a significant modulator of immune/cancer cell communication [39]. Nevertheless, the diversity of EVs and their functions contribute to the complexity and ambiguity of their anti-tumor and pro-tumor effects (Fig. 3). Research into the biological functions of EVs has revealed their involvement in various cellular processes that support cancer development and therapeutic outcomes, showcasing both cancer-promoting and cancer-inhibiting characteristics. The immune function of EVs associated with tumors is likely intricate and dynamic, primarily functioning through intercellular communication. This includes processes such as natural killer (NK) cell activity, antigen presentation, modulation of immune responses, and the establishment of immune tolerance.

Fig. 3.

EVs function as a double-edged sword in the tumor immune environment. The intricate cell composition of the tumor microenvironment contributes to the complexity of the origin and destination of tumor-associated EVs, leading to a diverse range of immunosuppressive and immune activation effects. Tumor cells, immune cells, and other relevant cells exhibit dual functions in both promoting tumor immunity and facilitating immune evasion

Transfer of antigens to antigen-presenting cells (APCs) through TEXs and immune cell-derived exosomes (IEXs) has been shown to activate antigen-specific T-cell responses, thereby enhancing the antitumor response. The presence of MHC class I and II molecules, as well as T cell co-stimulatory molecules on the surface of these EVs, plays a crucial role in antigen presentation.

However, EVs also exhibit immunosuppressive properties that can influence the immune response by transporting tumor antigens and immune checkpoint molecules [177]. They play a role in creating a pre-metastatic niche and fostering an immunosuppressive environment to aid tumors in evading immune surveillance [178]. Furthermore, EVs can engage with immune cell surface receptors to facilitate their uptake by immune cells, potentially enabling selective uptake of EVs containing specific antigens [179]. These findings suggest that EVs may play a crucial role in modulating tumor immunity, although further research is required to elucidate the exact mechanisms involved.

In the upcoming paragraphs, we will delve into the EV elaboration from diverse sources, including different tumor cells, immune cells, and other relevant cells found in the TME. These EVs play a crucial role in orchestrating the interaction among tumor immune cells by modulating processes such as NK cell activity, antigen presentation, cellular immunity, and humoral immunity within the immune system. This regulation ultimately aims to influence tumor immune responses and the microenvironment.

Tumor cell-derived EVs act on tumor immunity

TEXs play a crucial role in the immune response of tumor cells by modulating tumor immunomodulation through three key mechanisms: presenting antigens and immunosuppressive molecules, influencing interactions between tumor immune cells, and altering the TME. Tumor cells have been observed to evade the immune system by decreasing their immunogenicity, promoting the generation of suppressor cells, controlling antigen presentation, and releasing immunosuppressive factors.

Recent literature suggests that TEX has potent immunosuppressive effects. Prolonged exposure to TEXs can convert immune subpopulations to a pro-tumorigenic phenotype, leading to the formation of pre-metastatic niches and an immunosuppressive microenvironment. Cancer-derived EVs inhibit CD8⁺ T cell proliferation and activation, promote regulatory T cell and myeloid-derived suppressor cell expansion, suppress NK cell activity, interfere with monocyte differentiation [180], and express molecules like PD-L1 and TGF-β that mediate immunosuppression [181, 182]. These actions create a conducive environment for tumor growth and contribute to drug resistance in cancer therapy [183].

While EVs are commonly associated with promoting tumor immune escape and resistance, they can also have an anti-tumor effect by presenting tumor antigens, activating T cells, or delivering stress protein HSP70 to activate NK cells. Additionally, TEXs have shown promise as anticancer vaccines in animal models, stimulating antigen-specific T and B cell responses. This suggests that the immune system’s response to TEXs may depend on the specific environment in which they are encountered, influencing whether they stimulate or suppress immune activity [184].

From the perspective of innate immunity, tumor cell-derived EVs may contain immunosuppressive molecules such as TGF-β and IL-10, which can hinder the function of innate immune cells like macrophages and NK cells. This inhibition assists tumor cells in evading attacks from the immune system. Additionally, these EVs may possess molecules that regulate inflammation, impacting the inflammatory environment within the tumor and subsequently influencing immune cell behavior. The TME is characterized as an immune-promoting milieu with elevated pro-inflammatory signals during the initial stages of tumor formation, wherein immune cells typically display a pro-inflammatory phenotype. As tumor proliferation and metastasis advance, the TME progressively transitions to a hypoxic environment marked by low pH, reduced glucose concentration, elevated fatty acid levels, diminished amino acid availability, increased adenosine concentration, and high lactate levels. This chronic inflammatory state is implicated in tumor initiation and progression [185]. Notably, EVs secreted by tumor cells play a pivotal role in sustaining this environment. In patients with brain metastasis, levels of CEMIP are found to be elevated in both tumor tissues and EVs, which subsequently triggers endothelial cell branching and inflammation within the perivascular microenvironment [186].

Furthermore, EVs can transport chemokines and cytokines that attract immune cells to the tumor site. While this recruitment can sometimes enhance immune responses against the tumor, it can also be utilized to steer immune cells away from the tumor by providing information on immune evasion strategies, like immune antigen avoidance or molecular mechanisms of immune evasion. This enables tumor cells to successfully evade detection by innate immunity. For instance, EVs from cancer cells have been shown to suppress NK cell proliferation and cytotoxic function by reducing NKG2D levels. TEXs can also trigger an increase in NOS2 levels, inhibit mitochondrial oxidative phosphorylation, and boost glycolytic processes through NF-κB-dependent metabolic changes. This leads to elevated PD-L1 expression in macrophages, increased GLUT-1 levels in primary tumors, upregulation of YKT6 expression related to vesicle release, and enhanced infiltration of immunosuppressive macrophages in the pre-metastatic niche [187].

From the perspective of acquired immunity, TEXs carry tumor-specific antigens that are recognized by the immune system and trigger the acquired immune response. TEXs play a crucial role in facilitating the recognition and elimination of tumor cells by T cells and other immune cells. They achieve this by delivering tumor antigens to APCs, stimulating T cell responses, and potentially enhancing immune memory, thereby improving the immune system’s ability to respond to encountered tumor antigens [188]. On the other hand, TEXs can also dampen the immune response by transferring immune checkpoint molecules to T cells, leading to T cell exhaustion. In the context of immunotherapy, TEXs containing immune checkpoint-associated molecules may impact the efficacy of immune checkpoint inhibitors [188]. Additionally, studies have demonstrated that TEXs can trigger T-cell apoptosis through Fas ligand (FasL), inhibit T-cell receptor (TCR) activity, and alter the gene expression profiles of regulatory and effector T cells (Treg and Teff), with TGFB1 promoting the generation of Tregs [189]. The multifaceted role of EVs in various systemic tumors will be further explored in subsequent sections (Table 1).

Table 1.

Immunomodulatory effect of EVs derived from tumor cells

EV Source	Mediator	Related key molecules	Biofunctions	Ref.
NPC	Galectin-9	YAP1/ FPα pathway	Apoptosis of EBV-specific CD4 T cells, inhibition on Th1 cell function and reduce the infiltration of granzyme B-expressing CD8a+ T cells	[190, 191]
	miR-24-3p	FGF11	Inhibition of the differentiation of Th1 to Th17 immune cells	[192]
NSCLC	miR-21 and miR-29a	TLR8/NF-κB/IL-6/TNF-α	Increase in regulatory T-cell induction and inhibition of IFN-γ	[193]
Lung cancer	miR-3200-3p	DDB1	Promotes aging of Treg cells	[194]
Lung adenocarcinoma	TLR2	NF-kB /HIF-1α/GLUT-1 and NOS2/NO	Macrophage M2 polarization, up-regulated PD-L1 expression	[187]
Esophagus cancer	miR-21	STAT3	Promoting the formation of M-MDSCs	[195]
	O-GlcNAc transferase	PD-L1	Up-regulation of PD-L1 in CD8+ T cells	[195]
Gastric carcinoma	-	NF-κB	Promoting the release of pro-inflammatory cytokines and maintaining the tumor microenvironment in an inflammatory state	[196]
HCC	circCCAR1	miR-127-5p/WTAP	Enhancing the stability of PD-L1 in CD8+ T cells and contribute to immunosuppression	[197]
	circUSP7	miR-934/SHP2		[198]
	circMET	miR-30-5p/Snail/DPP4/CXCL1		[199]
	circUHRF1	miR-449c-5p/TIM-3	Inhibition of NK cell function, resulting in anti-PD-L1 resistance	[200]
Pancreatic cancer	FGD5-AS1	STAT3/NF-κB	Macrophage M2 polarization	[201, 202]
Colorectal cancer	PD-L1	TGF-β/Smad/SAPK	Increasing the proportion of PD-1+ Treg cells	[203]
	miR-208b	PDCD4	Promoting Treg amplification	[204]
Clear cell renal cell carcinoma		TGF-β/SMAD	Negative regulation of NK cells	[205]
Bladder cancer	CircRNA_0013936	miR-320a/JAK2 and miR-301b/CREB1	Promoting the expression of FATP2 in PMN-MDSCs and inhibiting the expression of RIPK3 significantly suppresses the function of CD8+ T cells	[206]
PCa	IL-8, PD-L1, NKG2D	PPARα and UCP1	Affecting the activity of CD8+ T cells	[207, 208]
Ovarian cancer	miR-940	CD163 and CD206	Macrophage M2 polarization	[209]
	miR-29a-3p	FOXO3-AKT/GSK3B	Enhancing the expression of PD-L1 in ovarian cancer cells	[210]
Cervical cancer	miR-1468-5p	HMBOX1/JAK2/STAT3	Increasing PD-L1 expression on lymphatic vessels and lymphangiogenesis, thereby weakening the immune response of T cells	[211]
	miR-27a-3p	MAGI2/PTEN/PI3K	Upregulating macrophage PD-L1	[212]
Glioma	EIF2, mTOR and ephrinB	PD-L1 and STAT3	Macrophage M2 polarization	[213]
	miR-1246	DUSP3/ ERK and STAT3/NF-κB	Driving the differentiation and activation of MDSCs, promoting the M2 polarization of macrophages	[214]
	miR-10a and miR-21	PTEN and RORA	Expansion and activation of MDSCs	[215]
	-	IFN-g and Granzyme B	Suppressive function on the cytotoxic activity of CD8+ T cells	[216]
	IgG	-	Inducing normal monocytes to differentiate into Th2 monocytes	[217]
B-cell lymphoma	HSP60, HSP90, MHC I, MHC II, CD40, CD86, RANTES and IL-1β	NQO1, PD1 and FoxP3	Release of specific cytokines with immunosuppressive functions, promoting the proliferation of T cells	[218]
CLL	noncoding Y RNA hY4	TLR7	Monocytes and macrophages tend to support tumor-promoting phenotypes, overexpression of genes related to suppressive T cells	[219]
Melanoma	-	LEC uptake	Cross-presentation on MHC-I and induction of antigen-specific CD8+ T cell apoptosis	[220]

EVs derived from nasopharyngeal carcinoma (NPC) cells have been shown to induce apoptosis in a significant number of Epstein-Barr virus (EBV)-specific CD4 + T cells and inhibit the function of Th1 cells through the presence of galactoglucan lectin 9 (Gal-9) [190]. Furthermore, EBV-positive NPC cells were found to reduce the infiltration of granzyme B-expressing CD8a⁺ T lymphocytes via activation of the YAP1/FAPα pathway [191]. Additionally, EVs present in the blood of nasopharyngeal carcinoma patients exhibit elevated levels of miR-24-3p, a molecule that hinders T cell proliferation and the differentiation of Th1 and Th17 cells by targeting fibroblast growth factor 11 (FGF11) [192]. In the context of non-small cell lung cancer (NSCLC), tumor cells release miR-21 and miR-29a into EVs, which are subsequently transported to the TME by TAMs. This transfer facilitates the binding of these microRNAs to Toll-like receptor 8 (TLR8). The resultant interaction activates the NF-κB signaling pathway, leading to an increased production of IL-6 and tumor necrosis factor-α (TNF-α). This cascade ultimately fosters a pro-inflammatory environment that promotes the proliferation and metastasis of cancer cells [193]. EVs derived from the blood of NSCLC patients carry PD-L1, which inhibits the generation of small po-factors and significantly reduces the secretion of interferon-gamma (IFN-γ) by Jurkat T-cells [193]. Furthermore, VEGFR2 can stimulate the upregulation of miR-3200-3p expression in lung cancer EVs, which is subsequently delivered to the TME and targets DDB1 in Treg cells to induce Treg cell senescence, thereby impeding tumor advancement [194]. Lung adenocarcinoma cell EVs are capable of triggering NF-kB activation through TLR2, utilizing HIF-1α/GLUT-1 to enhance glucose uptake in macrophages and NOS2/NO to suppress mitochondrial oxidative phosphorylation, promoting glycolysis, M2 macrophage polarization, and the upregulation of PD-L1 expression [187].

Esophageal cancer-derived EV-miR-21 activates STAT3 signaling to induce the generation of M-MDSCs, contributing to cisplatin resistance in esophageal squamous cell carcinoma (ESCC) [195]. Extracellular vesicular O-GlcNAc transferase (OGT) from esophageal cancer stem cells promotes cancer cell growth and metastasis by upregulating PD-L1 in CD8⁺ T cells [195]. EVs derived from gastric cancer cells activate the NF-κB signaling pathway in macrophages, leading to the release of pro-inflammatory cytokines that sustain an inflammatory TME, thus accelerating gastric cancer progression [196]. EV-derived circCCAR1 from HCC cells contributes to immunosuppression by enhancing PD-L1 stability in CD8 + T cells through the miR-127-5p/WTAP feedback loop, promoting HCC growth and metastasis [197]. Other similar mechanisms involve circUSP7/miR-934/SHP2 axis [198] and circMET/miR-30-5p/Snail/dipeptidyl peptidase 4 (DPP4)/CXCL1 axis [199]. Additionally, circUHRF1 inhibits NK cell function by degrading miR-449c-5p, leading to TIM-3 upregulation and anti-PD-L1 resistance [200]. Pancreatic cancer cells also release EVs, such as FGD5-AS1, which suppress cancer cell immune function by inducing M2 polarization of macrophages via the STAT3/NF-κB pathway [201, 202]. TEXs in mice colon cancer MC38 cell line induce systemic immunosuppression through PD-L1, resisting anti-PD-L1 antibody blockade and increasing PD-1⁺ Treg cells proportion. This effect can be mitigated by Rab27a and Coro1a knockdown [221], as well as by activating the TGF-β/SMAD signaling pathway, inactivating SAPK signaling, and upregulating Tregs-related genes [203]. EV-miR-208b from colon cancer cells can also enhance Treg amplification by targeting programmed cell death factor 4 (PDCD4) when delivered to T cells [204].

Wu et al. found that EVs derived from clear cell renal cell carcinoma patients could downregulate tumor-infiltrating NK cells by activating the TGF-β/SMAD signaling pathway, indicating a potential impact of these EVs on immune cell function in the TME [205]. Furthermore, circRNA_0013936 from bladder cancer EVs was shown to directly target JAK2 and CREB1 through miR-320a and miR-301b, leading to increased FATP2 expression and decreased RIPK3 expression in PMN-MDSCs, resulting in significant suppression of CD8⁺ T cell function [206]. Prostate cancer-derived EVs were also found to modulate immune responses by activating PPARα and UCP1 in CD8⁺ T cells through IL-8 translocation, which reduces glucose utilization, enhances fatty acid catabolism, and inhibits CD8 + T cell activity [222]. Moreover, PD-L1 and NKG2D present in prostate cancer EVs were shown to impact CD8 + T cell function in a similar manner [207, 208].

The majority of studies on EVs derived from reproductive tumors have primarily centered around ovarian cancer. EVs from ovarian cancer cells have been found to carry miRNAs like miR-940 into the TME, leading to M2 polarization in macrophages [209]. Additionally, miR-29a-3p targets the FOXO3-AKT/GSK3B axis, enhancing PD-L1 expression in ovarian cancer cells [210]. This increase in PD-L1 expression facilitates immune evasion by ovarian cancer cells, contributing to tumor progression. Breast cancer-derived EVs play a crucial role in suppressing immune function to facilitate tumor progression. Extracellular vesicles released from different breast cancer subtypes exhibit varying immunomodulatory effects. Specifically, EVs from triple-negative breast cancer (TNBC) patients possess potent immunosuppressive properties, leading to a reduction in CD3⁺HLA-DR⁺ T cells and an increase in CD3⁺ PD-L1 T cells. This is accompanied by an elevation in CD4 + CD127-CD25^hi Tregs, IL-10 levels, and enhanced production of anti-inflammatory cytokines like IL-10 [223]. Additionally, breast cancer-derived EVs have the ability to impede the cytotoxicity of NK cells, thereby enabling immune evasion [224]. In cervical cancer, cancer-associated lymphatic endothelial cells (LEC) present a significant obstacle to anti-tumor immune responses. The mechanisms through which they exert immunosuppressive effects in the TME remain incompletely understood. Recent research suggests that miR-1468-5p from cervical cancer EVs plays a role in increasing PD-L1 expression in lymphatic vessels and lymphatic neovascularization, thereby hindering T cell-mediated immune responses. Moreover, miR-1468-5p within EVs epigenetically activates the JAK2/STAT3 signaling pathway in LEC by targeting homology box inclusion 1 (HMBOX1) in the SOCS1 promoter, initiating an immunosuppressive program that enables cancer cells to evade antitumor immunity [211]. Additionally, endoplasmic reticulum stress promotes the secretion of cervical cancer EVs, leading to elevated expression of miR-27a-3p in EVs. This miRNA can upregulate PD-L1 by targeting and negatively regulating the MAGI2/PTEN/PI3K axis in macrophages [212].

EVs derived from glioblastoma cells are critical in accelerating glioblastoma progression by upregulating PD-L1 expression, activating STAT3, and inducing macrophage polarization towards the M2 phenotype, thereby facilitating immune escape [213]. EVs found in the cerebrospinal fluid of glioma patients are notably enriched with miR-1246, which drives the differentiation and activation of MDSCs through a DUSP3/ERK-dependent mechanism. Under hypoxic conditions, EVs exhibit higher levels of miR-1246, serving as a pro-oncogenic factor that enhances cancer progression by modulating STAT3 activity, inhibiting the NF-κB pathway, and promoting M2 polarization of macrophages [214]. Furthermore, glioma cell-derived EVs enriched with miRNA-10a and miRNA-21 lead to expansion and activation of MDSCs, impairing immune function by targeting PTEN and RORA [215]. Experimental evidence using the mouse glioma cell line GL26 confirmed the inhibitory effect of glioma-derived EVs on CD8⁺ T cell cytotoxic activity [216]. Additionally, EVs present in the blood of patients diminish the cytotoxic function of CD4⁺ T cells, favoring a Th2 response and aiding glioblastomas in evading immune surveillance [217].

In various hematologic malignancies, immune cells often transition to an immunosuppressive state. B-cell lymphoma cells release EVs carrying CD20, which can bind therapeutic anti-CD20 antibodies, deplete complement, and shield target cells from antibody attacks [225]. Additionally, these EVs from B lymphomas are abundant in heat shock proteins like HSP60 and HSP90, along with other immunogenic molecules like MHC I, MHC II, CD40, CD86, RANTES, and IL-1β [226]. Research has shown that B lymphoma cells can stimulate the proliferation of numerous T-cell clones by releasing EVs, which in turn promote the efficacy of a DC-derived EV vaccine. This interaction between EVs and DCs leads to the proliferation of T-cell clones and the modulation of immune responses through the release of specific cytokines like IL-14 and IL-10 with immunosuppressive properties [227]. Moreover, the interaction between DCs and B lymphoma cells results in the increased release of inflammatory factors such as IL-6 and TNF-α. NKG2D, an activation receptor for NK, NKT, CD8⁺, and gamma delta⁺ T cells, plays a crucial role in immune evasion when aberrantly lost in cancer [228]. Membrane-associated TGF-β1, MICA/MICB, and myeloid progenitor cell markers CD34, CD33, and CD117 have been identified in the sera of patients with acute myeloid leukemia. These components have been shown to decrease the expression of the NKG2D receptor on normal NK cells, leading to a significant reduction in NK cell cytotoxicity [159]. Similar effects have been observed in other cancer cell lines, such as prostate cancer [182]. Additionally, plasma EVs from patients with chronic lymphocytic leukemia contain high levels of noncoding Y RNA hY4, which activates TLR7 signaling and promotes a pro-tumorigenic phenotype in monocytes and macrophages. This phenotype is characterized by the secretion of cytokines like CCL2, CCL4, and interleukin 6, as well as the upregulation of PD-L1 expression [219]. Furthermore, exposure to EVs from K562 cells in chronic granulocytic leukemia leads to a dose-dependent differentiation of macrophages into TAMs [229]. This differentiation is accompanied by the overexpression of suppressor T cell-related genes such as NQO1, PD1, and FoxP3, resulting in enhanced immunosuppression [218].

Plebanek et al. demonstrated that EVs released by melanoma cells can activate the innate immune response by recruiting NK cells and TRAIL-dependent killing, initiating immune surveillance, and activating the body’s intrinsic immunity. This ultimately resulted in the eradication of tumor cells before metastasis and effectively suppressed lung metastasis [230]. Tumor-draining lymph nodes (LNs) play a crucial role in tumor progression, often serving as the initial site of metastasis with significant predictive value across various cancer types. During primary tumor growth, these LNs undergo extensive remodeling characterized by leukocyte infiltration and local cell proliferation, including lymph node-resident macrophages and stromal cells such as LEC. Injection of melanoma-derived EVs revealed their uptake by LN-resident LEC, leading to cross-presentation on MHC-I and subsequent induction of antigen-specific CD8⁺ T cell apoptosis [220].

As a significant therapeutic modality for localized tumors, radiotherapy induces cytoplasmic accumulation of double-stranded DNA (dsDNA) in cancer cells, leading to DNA damage response. This activation of type I interferon (IFN-I) via the cGAS/STING pathway enhances immunogenicity in irradiated cancer cells and facilitates the metastasis of tumor antigens [231, 232]. Additionally, it promotes tumor antigen transfer, triggers dendritic cell (DC) maturation, initiates effector T cell growth, and increases NK cell numbers. This shift in the TME towards an immunostimulatory state results in the recruitment of more T cells. Systemic antitumor effects, known as distal effects, are observed when tumors within the irradiated field regress, although these effects are not yet fully understood. The increased release and modified content of EVs from donor cells post-irradiation may contribute to these distal effects [233–235]. Research has demonstrated that the molecular composition of radiation-induced TEX (RT-TEX) from TSA mouse mammary carcinoma cells irradiated with 8 Gy X 3 differs significantly from TEX of untreated cells [82]. RT-TEX has been shown to activate DC cells and enhance the infiltration of CD4⁺ and CD8⁺ T cells within tumors in in vivo experiments [82, 236]. In renal cell carcinoma cell lines, Sunitinib-resistant (SR) cell-derived EVs had higher expression of PD-L1 on their surface compared to Sunitinib-sensitive (SS) cell-derived EVs. This led to increased cytotoxicity towards immune cells, a phenomenon linked to elevated levels of PD-L1 protein on SR EVs. Conversely, Tipifarnib was shown to inhibit EV formation and release, as well as reduce PD-L1 expression on both SS and SR cells, offering potential clinical benefits for patients [237]. Furthermore, in patients undergoing anti-PD-1 therapy for metastatic melanoma, levels of EV PD-L1 correlated with IFN-γ and changed with treatment progression. Elevated EV PD-L1 levels at treatment initiation indicated a tumor cell response to T-cell attack, aiding in distinguishing between responsive and nonresponsive patients [238]. Heat stress-treated EVs (HS-TEX) administered in vivo to tumor patients undergoing thermotherapy were found to enhance the conversion of T regs to Th17 cells by upregulating IL-6 and IL-17 levels. Additionally, they inhibited the differentiation of Tregs induced by tumor cells, thereby exerting a tumor-suppressive effect. The observed elevation in serum IL-6 and IL-17 levels in tumor patients receiving thermotherapy, along with the increase in Th17 cells and decrease in Tregs in peripheral blood mononuclear cells, provide further evidence of the role of HS-TEX in promoting an effective anti-tumor immune response [239] (Table 2).

Table 2.

Immunomodulatory effects of EVs under the influence of external factors

External Factors	EV Source	Mediator	Target Cells	Related key molecules	Type of Diseases	Biofunctions	Ref.
Radiotherapy	Mouse breast cancer cells	dsDNA	DC and T cell	cGAS/STING	breast cancer	Activating DC cells and enhancing the infiltration of CD4+ T cells and CD8+ T cells in the tumor	[82, 236]
Chemotherapy resistance(Sunitinib)	Renal cell carcinoma cell	PD-L1	SS and SR cells	-	renal cell carcinoma	More obvious cytotoxicity to immune cells	[237]
Anti-PD-1 therapy	Metastatic melanoma cells	PD-L1	T cells	-	melanoma	Inhibition of CD8+ T cell activity	[238]
Thermal therapy	MC38 cells	IL-6 and IL-17	Th17 and Treg	-	Malignant ascites in colorectal cancer patients	Th17 cells are increased and Tregs cells are decreased	[239]
Anoxia	glioma cell	miR-1246	macrophage	STAT3/ NF-κB	glioma	Promoting macrophage M2 type polarization	[214]
		miRNA-10a and miRNA-21	MDSC	PTEN and RORA		Amplification and activation of MDSC	[215]
Respiratory syncytial virus infection	lung carcinoma cell A549	Small ncRNA	human monocytes and airway epithelial cells	-	lung cancer	Human monocytes and airway epithelial cells are induced to release cytokines and chemokines to activate the innate immune response	[240]

Innate immune-cell-derived EVs act on tumor immunity

Immune cells release chemokines via EVs, attracting other immune cells to migrate towards the tumor region. This process is crucial for local immune responses, regulating immune cell distribution, and facilitating information exchange in the TME. EVs secreted by immune cells exhibit immunomodulatory functions similar to immune cells, with variations based on cell type. Innate immune cell-derived EVs play a pivotal role in tumor immunity regulation by influencing interactions between immune cells and the TME. Cuturi et al. [241] demonstrated that EVs from bone marrow dendritic cells can impact allograft rejection by presenting donor major histocompatibility complex antigens, showcasing the immune-modulating potential of EVs. It is important to recognize that EVs from innate immune cells can modulate immune responses to tumors in diverse ways, some supporting immune evasion while others enhancing immune system efficiency in combating tumors. The following discussion will explore these possibilities, beginning with different innate immune cell contributions (Table 3).

Table 3.

Immunomodulatory effect of EVs derived from innate immune cells

EV Source	Mediator	Target Cells	Related key molecules	Biofunctions	Ref.
MCs	CD63 and OX40L	T cells	OX40	Promoting Th2 cell proliferation and differentiation	[242]
DCs	MHC-II, ICAM-1	T cells	-	Inducing a strong adaptive immune response	[243]
	OX40L	T cells	IFN-γ and IL-4	Promoting CD4+ T cell proliferation and IL-4 production, and guiding Th2 differentiation	[244]
	-	SK-BR-3 breast cancer cell	IFN-γ	Increasing IFN-γ production from CD3+ T cells	[245]
	HSP70 and HSP90	T cells	MHC-I、MHC-II和CD86	Activating CD4+ and CD8+ T cells	[246]
	TNF-α	B cells	NF-κB	Inducing inflammatory response	[247]
TAMs	miR-21-5p and miR-29-3p	T cells	STAT3	Inducing Treg/Th17 imbalance and promoting the progression and metastasis of epithelial ovarian cancer cells	[248]
	PD-L1	T cells	-	Inhibiting the proliferation and activity of CD8+ T cells	[171]
	LINC01592	cancer cells	E2F6/NBR1/MHC-I	Allowing cancer cells to evade immune attack from CD8+ T cells	[249]
	miR-21	T cells	PEG3	Accelerating immune escape from glioma	[250]
MDSCs	miR-29a-3p and miR-93-5p	T cells	T-bet and STAT3	Inhibit Th1 and Th17 cell differentiation	[251]
	Fas and TNF-1α	T cells and macrophage	IFN-γ and ROS	Reducing the proportion of M1 macrophages and depletion of CD8+ T cells	[252]
Neutrophil	miR-30d-5p	macrophage	NF-κB	Increasing polarization ratio of M1 macrophages and pyroptosis of macrophages	[253]
NKs	miR-186	Neuroblastoma cell line	TGFB1	Inhibition of TGFB1-dependent immune escape	[254]

Mast cells play a crucial role in both early and late IgE-mediated allergic inflammatory responses. Activation of these cells occurs when FcεRI cross-links bound IgE with multivalent antigens, leading to the release of pre-formed mediators, newly synthesized lipid mediators, and cytokines. The diverse biological effects of these mediators are responsible for the central role of mast cells in acute allergic reactions and chronic allergic inflammation. Beyond IgE-mediated reactions, mast cells are also involved in various inflammatory responses. Studies have shown that mast cells or their products are essential in recruiting leukocytes to sites of inflammation, presenting antigens to T cells, interacting with cells of the adaptive immune system, and mediating tissue remodeling [255]. Research has demonstrated that bone marrow-derived mast cells release EVs expressing CD63 and OX40L, with OX40L promoting the proliferation and differentiation of Th2 cells by binding to OX40 on T cells [242]. EVs derived from the mast cell line HMC-1 can be internalized by lung cancer cells, specifically A549. This uptake facilitates the transfer of KIT protein, which subsequently activates the KIT-SCF signaling pathway. As a result, there is an upregulation of cyclin D1 expression, leading to enhanced proliferation of human lung adenocarcinoma cells [256].

DCs are a crucial class of APCs responsible for capturing antigens and processing them into peptides. These peptides are then presented to T cells through MHC molecules on the cell surface, initiating an adaptive immune response. Serving as essential mediators between the innate and adaptive immune systems, DCs express a diverse array of TLRs and cytokines that actively participate in immune response activation. Research has highlighted the significance of TLR4 expression in driving the anti-cancer effects observed in DC-based immunotherapy. Moreover, DCs, originating from monocytes, produce IL-12 to facilitate the differentiation of Th1 cells from initial CD4⁺ T cells. With their specialized role as antigen-presenting cells, dendritic cells possess the unique capability to elicit both primary and secondary immune responses [257]. DCs have a unique ability to trigger immune responses through the interaction of DC-derived EVs (DEX) with specific cell types. These EVs prompt the release of immune-activating molecules like cytokines, leading to the activation of various immune cells such as macrophages, NK cells, and neutrophils. In addition to carrying typical EV marker proteins CD9, CD63, and CD81, DEX also contain important molecules for cell-to-cell interactions, including MHC molecules, co-stimulatory molecules (CD80, CD86, and CD40), and integrins [258]. By presenting antigenic peptides in both direct and indirect ways, DEX play a crucial role in promoting a robust immune response to cancer cells. Direct presentation involves DEX directly presenting antigens to immune cells, educating them to respond more aggressively to invaders. Indirect presentation occurs when DEX are internalized and expressed by DCs, facilitating the presentation of antigens to immune cells. This process involves multiple cells simultaneously and enhances the immune response against cancer cells [259].

Under various conditions, DEX secretion exhibits comparable immune-activating properties. EVs derived from IFN-γ-stimulated DCs play a crucial role in initiating a robust adaptive immune response. These EVs present class II peptide complexes to initial T lymphocytes, leading to specific cytotoxic T-cell responses in vivo, which have been shown to effectively suppress or eliminate tumors in mouse models [260, 261]. Additionally, thymic stromal lymphopoietin (TSLP)-induced EVs released by DCs contain OX40L, a protein that enhances CD4⁺ T-cell proliferation, IL-4 production, and guides Th2 T-cell differentiation [244]. DEX has been found to enhance IFN-γ production by CD3⁺ T-cells during interactions with SK-BR-3 breast cancer cells [245]. Furthermore, DCs release different types of EVs based on their composition, such as HSP70 and HSP90. These chaperone proteins activate CD4⁺ and CD8⁺ T cells, as well as MHC-I, MHC-II, and CD86 [246]. In addition to inducing the T-cell immune response, DEX also triggers the innate immune response by activating NK cells through ligands for NKG2D-L and IL-15Rα [262]. Furthermore, DEX indirectly activates T cells through antigen-MHC combinations and by delivering ICAM-1 molecules to less efficient APCs like B cells [243]. The activated B cells then produce and release EVs that can effectively stimulate both CD4⁺ and CD8⁺ T cells, thereby initiating a robust tumor immune response [263]. Research has demonstrated that EVs derived from DCs have the ability to trigger specific immune responses by prompting the release of inflammatory markers. When DCs present alpha-fetoprotein (AFP) on EVs, it leads to the activation of CD8 + T cells through a sequence of events. This process begins with a significant increase in the production of IFN-γ and IL-2, followed by the suppression of immunosuppressive factors in T cells such as CD25⁺Foxp3⁺ Tregs, IL-10, and TGF-β [264]. Additionally, EVs containing TNF-α released by DCs exhibit a strong interaction with NF-κB on B cells, which enhances an inflammatory reaction in human umbilical vein endothelial cells [247]. Wang and colleagues explored the immune-activating potential of DC-derived EVs and developed a CD4⁺ T cell-based vaccine to directly trigger the activation of CD8 + T cells. This activation of CD8⁺ T cells by the CD4⁺ T cell vaccine led to the development of acquired immune memory. The research focused on the uptake of DC-derived EVs by CD4⁺ T cells, which contained the antigenic peptide MHC-I crucial for immune memory, along with internal IL-2 and endogenous CD40L signaling. Administration of a CD4⁺ T cell-based vaccine in mice demonstrated the capacity to induce a more robust CTL activation linked to CD8⁺ T cells [265]. While DC-derived EVs presenting antigenic MHC peptides tend to effectively activate immune responses, some uncertainty remains. Notably, in certain experiments, MHCI^−/− mice treated with ovalbumin-loaded dendritic cell-derived EVs displayed a similar induction of antigen-specific T cells compared to wild-type EVs. This suggests that immune activation may occur independently of DC-based EV pMHC if the entire antigen is present [266].

Macrophages, derived from bone marrow, are crucial immune cells involved in both non-specific and specific immune responses in the body. These cells can differentiate into either pro-inflammatory M1 type or immunosuppressive M2 type based on stimuli from cytokines. M1-type macrophages, activated by bacterial products like LPS, primarily engage in phagocytosis of pathogens and cellular debris, and initiate inflammatory responses by secreting pro-inflammatory factors such as IFN-γ, TNF-α, and IL-6. On the other hand, M2-type macrophages, activated by IL-4 and IL-13, promote tissue repair through the secretion of IL-10 [267]. In nonspecific immunity, macrophages help regulate the inflammatory response by clearing dead cells, cellular debris, and foreign bodies. In specific immune responses, they release cytokines like IL-1, IL-8, IL-6, and TNF-α, and present antigens to stimulate an immune response [268]. The diverse origins of macrophages give them and their EVs varied functions. For instance, studies have shown that EVs from LPS-stimulated RAW264.7 macrophages [269] or MFGE8-containing Schwann cell EVs [270] can have neuroprotective effects by inducing microglia polarization towards the immunosuppressive M2 type. TAMs typically exhibit characteristics similar to M2 macrophages and play a crucial role in the TME. They are known for their lack of cytotoxic activity, provision of growth factors to cancer cells, and immunosuppressive functions. Recent studies have highlighted the involvement of TAM-derived EVs in facilitating immune evasion by tumor cells. EVs released by M2 macrophages contain miR-155-5p, which hinders ZC3H12B-mediated IL-6 stability, promoting immune escape and colon cancer progression. Furthermore, the delivery of miR-21-5p and miR-29-3p by TAM EVs contributes to the advancement and metastasis of epithelial ovarian cancer cells by disrupting Treg/Th17 balance through STAT3 inhibition [248]. Additionally, TAM EVs are rich in PD-L1, which effectively suppresses the proliferation and function of CD8⁺ T cells [171]. Analysis of melanoma tissues from patients has revealed the significant role of TAM EVs in CD8⁺ T cell inhibition [171], with CD163⁺ TAM-derived EVs being the primary mediators. Moreover, TAM-secreted EVs containing LINC01592 promote NBR1 gene transcription in tumor cells by binding directly to E2F6. Subsequently, NBR1 interacts with ubiquitinated MHC-I proteins, leading to increased degradation of MHC-I in autophagic lysosomes and reduced MHC-I expression on tumor cell surfaces, enabling evasion of CD8⁺ T cell immune responses. Targeting the E2F6/NBR1/MHC-I signaling pathway using siRNA or specific antibodies can significantly impede the tumor-promoting effects of LINC01592 and M2-TAM-driven tumor growth [249]. Similarly, EVs derived from bone marrow M2 macrophages can impede CD8⁺ T cell expansion and cytotoxicity through miR-21/PEG3, facilitating immune evasion in gliomas [250].

Myeloid-derived suppressor cells (MDSCs) are specialized pathological myeloid cells that exhibit potent immunosuppressive activity and are produced in various pathological conditions. These cells are categorized as granulocytic or polymorphonuclear cells (PMN-MDSC) and monocytes (M-MDSC) [271]. EVs released by MDSCs inherit the immunosuppressive functions of the parent cells and play a role in processes such as tumor growth, angiogenesis, invasion, and metastasis. Studies have shown that MDSC-derived EVs contain miRNAs that can modulate the suppressive function of MDSCs by influencing myeloid cell differentiation and proliferation, inducing senescence in CD8⁺ T cells, and inhibiting the proliferation of CD8⁺ T cells [272]. Furthermore, these EVs hinder the differentiation of Th1 and Th17 cells by targeting T-bet and STAT3 through miR-29a-3p and miR-93-5p, respectively [251]. Treatment with MDSC-derived EVs also results in a significant decrease in M1 macrophages [252].

Neutrophils, originating from hematopoietic stem cells in the bone marrow through granulopoiesis, are activated and swiftly migrate from the bloodstream to infection or inflammation sites within around 30 min. Upon arrival, they undergo transformation into pus cells and eventually perish. Constituting about 60–65% of the total leukocyte population, neutrophils are the most abundant type of circulating leukocyte. As specialized phagocytes, they possess the ability to rapidly engulf and eliminate invading microorganisms and damaged cells, equipped with a variety of antimicrobial agents, enzymes, and toxic substances to aid in pathogen eradication. Neutrophils are vital components of the immune system, crucial for defending the body against infections by engulfing and destroying pathogens, and are indispensable for overall bodily functions [273]. A study found that miR-30d-5p in polymorphonuclear neutrophil EVs induces M1 macrophage polarization and triggers macrophage pyroptosis through the activation of NF-κB signaling [253].

NK cells were initially characterized in 1976 as effector lymphocytes of innate immunity, serving as a crucial defense mechanism against pathogenic microorganisms. Unlike T and B cells, NK cells do not rely on human leukocyte antigen molecules (HLA) for their cytotoxic function. This unique feature enables NK cells to play a pivotal role in combating malignancies, contributing to immune responses against both virally infected and tumor cells. Activation of NK cells can be triggered by cytokines such as IL-2, IL-12, IL-15, and IL-18, or by the detection of cells lacking MHC class I surface molecules. Upon activation, NK cells release cytotoxic particles that lead to the lysis or apoptosis of the target cell [274]. In addition to their essential functions in mediating the innate immune response, recognizing infected cells deficient in class I MHC proteins, and exerting antitumor effects, NK cells can also influence the immune response through the release of EVs. A study on cancer patients revealed that NK cell-derived EVs exhibited signature markers of NK cells, such as CD56, NKG2D, CD94, CDL40, as well as killer proteins like FasL, perforin, or granzyme [275]. These EVs directly target cancer cells, impede their migration to the tumor site, and primarily interact with cancerous tissues, showcasing the cytotoxic effects of NK cells [276]. Furthermore, EVs released by NK cells previously exposed to neuroblastoma (NB) have the ability to educate unexposed NK cells to demonstrate a more potent and effective cytotoxic response against NB tumor cells. Specifically in neuroblastoma, NK cell-derived EVs containing the tumor suppressor miR-186 can inhibit TGFB1-dependent immune evasion and reduce the tumorigenic potential [254].

Acquired immune-cell-derived EVs act on tumor immunity

The regulation of adaptive immunity by innate immune cells is a well-established concept. Huang et al. demonstrated that B cell-derived IL-10 plays a role in driving macrophage polarization towards the M2 type in a melanoma tumor model [277]. Additionally, EVs from activated CD3⁺ T cells have been shown to stimulate the proliferation of resting CD3⁺ T cells [278]. This suggests that EVs derived from acquired immune-associated cells can influence the immune response to tumors through various mechanisms. In the following sections, we will discuss how EVs from different acquired immune cells can modulate the immune response to tumors [279] (Table 4).

Table 4.

Immunomodulatory effect of EVs derived from acquired immune cells

EV Source	Mediator	Target Cells	Related key molecules	Biofunctions	Ref.
CLT	RAS/MAPK	MCs	IL-24 and ERK	Activation of ERK phosphorylation in MCs	[280]
Th cells	LAMP-1, TCR, and LFA-1; CD4, TCR, LFA-1, CD25, and Fas	CTL	-	Activating CD8+ T cells	[281]
Tregs	IL-35	T cells and B cells	Ebi3 and p35	Inducing peripheral immune tolerance	[282]
	CD39 and CD73	Th cells	adenosine	Inhibiting the activation and proliferation of CD4+ T cells	[283–285]
	Let-7b, Let-7d, miR-155, miR-142-3p and miR-150-5p	Th cells	-	Inhibiting Th1 immune response and strengthen the inhibitory function of the immune system	[286]
	miR-150-5p and miR-142-3p	DC	-	Decreasing secretion of IL-10 and IL-6	[287]
B cells	pMHC-II	T cells	-	Activating T cells	[288]

In acquired immunity, T cells are derived from bone marrow progenitors, undergo maturation in the thymus, and are subsequently released into the peripheral system. These cells play a crucial role in specific anti-tumor immunity. T cells can be broadly classified into Helper T cells (Th) and Cytotoxic T cells (Tc) based on their roles in the immune response and surface molecules. Helper T cells are identified as CD4 + T cells, while CTL are known as CD8 + T cells. The differentiation of these cells is influenced by the co-stimulation of MHC class I molecules presenting antigens on APCs and CD80 or CD86 on the same APCs. Upon activation, IL-2 promotes the proliferation of CTL [289].

EVs from CD4⁺ and CD8⁺ T cells have been shown to induce DC apoptosis and T cell suppression through a unique molecular interaction involving peptide/MHC/TCR and ICAM-1/LFA-1 [290, 291]. Additionally, mast cells have been found to uptake T cell derived EVs within 24 h, leading to increased cytokine secretion, including IL-24. Proteomics analysis has revealed that these EVs are rich in RAS/MAPK signaling proteins, which in turn trigger ERK phosphorylation in mast cells [280]. The capacity of CTL to release EVs with anti-cancer properties was illustrated in a mouse model of melanoma, where the spread and metastasis of tumors were notably diminished by the interference of cells within the tumor [292]. A robust immune response by CD8⁺ T-cells requires various components including antigenic, costimulatory, and inflammatory factors, such as IL-12. Research by Li et al. demonstrated that stimulating IL-12 to induce EV release activates the CD8⁺ T cell bystander effect, thereby improving the efficacy of immunotherapy by promoting the production of immune molecules like granzyme B (GZB) and IFN-γ [293].

CD4 acts as a co-receptor for the TCR on Th cells, enhancing their binding to MHC class II molecules on antigen-presenting cells. This process promotes the activation and differentiation of T helper cells. EVs released by CD4 helper T cells are enriched with characteristic EV markers such as LAMP-1, TCR, and LFA-1, along with CD4-specific markers like CD4, TCR, LFA-1, CD25, and FasL. These EVs have been demonstrated to enhance antitumor responses by activating CTLs [281].

Most Tregs are a subset of CD4 T cells with immunosuppressive functions, identified by the cell surface receptor CD25 and high levels of the transcription factor forkhead box P3 (Foxp3). Tregs play a crucial role in modulating the immune system and promoting tolerance to self-antigens [294, 295]. Recent studies have indicated that Tregs secrete EVs with a higher concentration of membrane molecules compared to other T cell subsets [286]. These Treg-derived EVs are able to achieve similar immunosuppressive effects in target cells as Treg cells themselves by delivering immunosuppressive molecules [290, 296]. Interleukin-35 (IL-35) plays a crucial role in inhibiting T cell activation, and the population of IL-35-rich Treg cells is elevated in the TME. IL-35 is released by Treg cells upon activation by the T cell receptor to restrict cellular infiltration and promote T cell depletion in the TME. EV IL-35 has the ability to specifically target T and B cells, inducing peripheral tolerance [282]. Furthermore, EVs generated by Treg cells expressing CD39 and CD73 contribute to the inhibition of CD4 + T cell activation and proliferation by generating adenosine [283–285]. EVs released by Treg cells contain miRNAs, including Let-7b, Let-7d, microRNA-155, miR-142-3p, and miR-150-5p. These miRNAs play a crucial role in inhibiting Th1-type immune responses by suppressing the translation of target mRNA molecules, thereby enhancing the suppressive function of the immune system. Treg cells deliver let-7d via EVs to Th1 cells, leading to the inhibition of Th1 cell proliferation and cytokine secretion, which helps in preventing systemic diseases [286]. Additionally, EVs enriched with specific miRNAs like miR-150-5p and miR-142-3p, when treated with LPS-stimulated DCs, reduce the secretion of IL-10 and IL-6 [287].

Unlike T cells, B cell differentiation and maturation take place in the bone marrow. B cells have crucial roles in host immunity, including producing immunoglobulins, presenting antigens, providing co-stimulatory signals, and releasing cytokines to modulate anti-tumor immunity [297]. The interaction between B cells and T cells, specific to antigens, is vital for an effective immune response. This interaction involves peptide-MHC class II complexes (pMHC-II) on B cells’ surface and TCRs on antigen-specific T cells. EVs released by B cells can activate specific T cells in humans and mice due to the presence of pMHC and co-stimulatory molecules. Approximately 12% of pMHC-II is released by B cells in EVs daily, which can activate T cells by binding pMHC-II to CD4 + T cells’ TCR [288]. In general, EVs originating from immune cells play a significant role in facilitating communication between innate and adaptive immune cells and serving as a link between the immune response and tumor cells.

EVs from other cellular sources act on tumor immunity

Tumor-associated fibroblast-derived EVs

The interaction between tumor cells and the surrounding stroma plays a crucial role in tumor development. Fibroblasts, which are cells originating from mesenchyme, produce components of the extracellular matrix that form the tumor stroma. Fibroblasts found in the stroma of cancer, commonly known as CAFs, contribute to the creation of the TME. Tumors have the ability to transform neighboring fibroblasts into CAFs. Activated CAFs can enhance tumor growth, angiogenesis, invasion, metastasis, extracellular matrix (ECM) remodeling, and even resistance to chemotherapy by establishing an optimized and immune-diverse microenvironment within the tumor.

Numerous studies have demonstrated that CAFs can influence tumor development by transmitting signals bidirectionally to tumor cells through EVs [298]. These CAF-derived EVs often contain death receptor ligands such as PD-L1 and inhibitory cytokines like TGF-β [299]. Specifically, EVs secreted by CAFs from breast tumors exhibit high levels of miR-92, which targets LATS2 and enhances the nuclear translocation of YAP2. This, in turn, facilitates the interaction between YAP1 and PD-L1 enhancer regions, ultimately promoting PD-L1 expression in breast cancer cells [300].

While the impact of CAF-derived EVs on immune cells within the TME remains largely unexplored, given the known role of CAFs in modulating the tumor immune response and their association with EVs, it is hypothesized that CAFs play a significant role in tumor immunomodulation mediated through EVs. Further research in this area is anticipated to uncover novel targets for future tumor prognostic evaluation and the development of therapeutic interventions.

Mesenchymal stem cell-derived EVs

MSCs are pluripotent and self-renewing stromal cells that can be found in various adult tissues. Initially discovered in bone marrow, subsequent research has revealed the presence of MSCs in adipose tissue, umbilical cord blood, placenta, and dental pulp. Multiple studies have demonstrated the significant and diverse immunomodulatory properties of MSCs, which impact both the innate and adaptive immune systems. In the innate immune system, allogeneic MSCs have been shown to hinder the maturation of monocytes or CD34⁺ haematopoietic precursor cell DCs, as well as their ability to release pro-inflammatory cytokines [301]. Additionally, allogeneic MSCs inhibit the IL-2-induced proliferation of resting NK cells, although their effect on activated NK cells is limited [302]. Interestingly, activated NK cells can lead to the destruction of MSCs [303].

Early studies have demonstrated that MSCs in the adaptive immune system possess direct immunosuppressive properties. These properties inhibit the activation and proliferation of effector T cells (CD4⁺ and CD8⁺) through cell-to-cell contact and the processing of various soluble factors such as IFN-γ, TNF-α, IL-10, PGE₂, HO-1, NO, HLA-G, MMP, chemokines, and adenosine. Additionally, MSCs can induce the generation and proliferation of T cell suppressor Tregs [304, 305]. MSCs derived from umbilical cord stroma, bone marrow, and adipose tissue sources have all been found to inhibit the proliferation of mitogen-activated T cells, induce an anti-inflammatory tolerance phenotype in effector T cells and NK cells, and inhibit the proliferation of B cells, with the exception of UCM-MSCs [306]. In summary, MSCs play a role in creating an immunosuppressive environment, a finding that was further supported by a study involving a mouse model of encephalomyelitis autoimmune disease (EAE) injected with BMSC [307].

MSCs exhibit remarkable migration towards tumor sites, a process known as homing. This migration is facilitated by various factors such as chemokines, growth factors, and extracellular matrix components within the TME. The recruitment of MSCs to tumors enables interactions with cancer cells and the microenvironment, thereby influencing tumor behavior. MSC-EVs, secreted by MSCs, display similar recruitment behavior and function as regulators of the tumor ecological niche. They achieve this by transferring EV contents, such as specific proteins or genetic material, to neighboring cancer cells and the microenvironment. This transfer process can either promote or inhibit tumorigenesis, angiogenesis, and metastasis depending on specific microenvironmental cues and recipient cells [308].

Additional research indicates that certain immunomodulatory characteristics of MSCs can be conveyed through the EVs they release. Specifically, a study demonstrated that hsa-miR-23b-3p present in Bone Marrow MSC-derived EVs plays a role in maintaining the balance between Th17 and Treg cells, thereby reducing the population of Th17 cells. This effect is achieved by targeting KLF5 to inhibit the PI3K/Akt/NF-κB signaling pathway [309]. On the other hand, EVs derived from Adipose tissue MSCs were shown to influence the function of CD11c⁺ DC cells by promoting the secretion of TGF-β and IL-10. This, in turn, led to an increase in the expression of DC co-stimulatory markers and enhanced the ability of DCs to regulate lymphocyte proliferation [310].

In conclusion, MSCs are considered an optimal source of EVs, which are believed to significantly influence a variety of MSC functions. MSC EVs have been extensively researched and are known for their ability to modify the phenotype or function of recipient cells through cell-to-cell communication. They represent a promising cell-free approach in place of MSC-based cell therapies [311].

Neural stem cell-derived EVs

Neural stem cells (NSCs) are self-renewing pluripotent cells with the ability to differentiate into neurons, astrocytes, and oligodendrocytes within the central nervous system [312]. In the context of CNS diseases, NSCs have emerged as a promising cell therapy approach [313, 314]. Their tumor-homing capabilities allow them to reach both primary and invasive tumor sites, presenting a unique delivery platform. Preclinical research has demonstrated that NSCs engineered with various cytotoxic agents can effectively target tumors, prolong the activity of anti-tumor treatments, reduce tumor size, and enhance survival rates [315]. Subsequent investigations have explored the potential of using NSCs as carriers for delivering a range of antitumor drugs, including chemotherapeutic agents, small molecule inhibitors, antibodies, cytokines, and lysoviruses, though not oligonucleotide therapy (ONT) [316]. In a study by Kortylewski M et al., EVs derived from NSCs loaded with antisense oligonucleotides (ASO) targeting STAT3 showed superior therapeutic effects against gliomas compared to natural NSC EVs. This was achieved by enhancing the immunoreactivity of human dendritic cells and mouse macrophages, indicating the potential of NSC EVs as effective drug carriers for tumor treatment [317].

The role of EV-related metabolic alterations in tumor immunity

From the above, we can conclude that EVs, which transport a diverse array of biomolecules including proteins, lipids, and nucleic acids, play a significant role in intercellular communication. EVs function not only as vehicles for information transfer but also as carriers of metabolic products and their regulatory enzymes. These biomolecules can reprogram various metabolic processes in receptor cells, including glucose, lipid, and amino acid metabolism, through multiple mechanisms. Additionally, they can promote the accumulation of metabolites such as lactic acid and fatty acids by modulating cellular metabolism within the TME. This alteration of the metabolic landscape influences tumor growth and subsequently regulates the energy metabolism and functional status of immune cells (Fig. 4). EVs may exert either immunosuppressive or immune-activating effects within the TME (Fig. 5). The following discussion will delve into the regulatory roles of EVs in tumor immunity through metabolic modulation.

Fig. 4.

Metabolic reprogramming in the tumor microenvironment. Metabolic reprogramming in the TME is a hallmark of cancer biology, where cancer cells and surrounding cells undergo significant metabolic changes to support tumor growth and survival. This process involves the redirection of nutrients and metabolic pathways to fuel the energy-demanding processes of cancer cells. (A) Schematic representation of glycolytic metabolic pathway. (B) Schematic representation of change of fatty acid oxidation. (C) Schematic representation of change of amino acid metabolism

Fig. 5.

EVs-mediated immune-related metabolic reprogramming in the tumor microenvironment. EVs play a pivotal role in the metabolic reprogramming of the tumor microenvironment, facilitating its remodeling through various mechanisms. They can transport metabolic enzymes, signaling molecules, and metabolic intermediates that alter the metabolic state of recipient cells, thereby influencing the behavior of immune cells, cancer cells, and stromal cells within the tumor microenvironment

Glycometabolism

Glucose metabolism encompasses the cellular processes that utilize and convert glucose, serving as a primary pathway for energy acquisition in cells. In tumor cells, this process often becomes dysregulated, leading to the phenomenon known as the Warburg effect. This effect is characterized by a preference for rapid ATP production via glycolysis, even in the presence of oxygen, rather than the more efficient oxidative phosphorylation pathway. Such a metabolic shift is crucial for meeting the energy demands associated with rapid proliferation and other functional needs of tumor cells [318]. The accumulation of lactate resulting from the Warburg effect, along with the acidification of the TME, plays a significant role in promoting tumor cell invasion, metastasis, and tumor-associated immune suppression. For example, TEXs from lung cancer can stimulate macrophages to enhance glucose uptake through the NF-κB pathway, which promotes glycolysis, elevates lactate levels, and increases PD-L1 expression, thereby polarizing macrophages toward an immunosuppressive phenotype [187]. Notably, high PD-L1 expression is positively correlated with the expression level of YKT6, a gene implicated in EV release [187]. Furthermore, lactate can also facilitate EV release in macrophages [319], and both the acidic TME and heightened EV release are recognized as critical phenotypes of malignancy [320]. This underscores the intricate interplay between EVs, metabolism, and metastasis within the TME.

Investigating the underlying mechanisms, EVs have been shown to directly transport proteins that play critical roles in glucose transport and metabolism, including phosphofructokinase (PFK) [321], transketolase, and transaldolase 1 [322]. Furthermore, mass spectrometry analyses of EVs derived from motile hepatocellular carcinoma cells (97 H and LM3) and non-motile Hep3B cells have identified distinct protein pathways associated with glucose metabolism, encompassing glycolysis, gluconeogenesis, and the pentose phosphate pathway [323]. These findings indicate that EVs can significantly influence glucose metabolism in recipient cells while simultaneously inhibiting immune cell function.

Furthermore, non-coding RNAs present in EVs can effectively target and regulate the expression of genes associated with glucose metabolism. For instance, lncMMPA found in EVs derived from TAMs in HCC enhances aerobic glycolysis through the miR-548s/ALDH1A3 pathway, which accelerates tumor progression and diminishes patient survival [324]. In pancreatic cancer, EV circPDK1 is activated by HIF1A under hypoxic conditions, promoting the proliferation, migration, and glycolysis of pancreatic cancer cells via the miR-628-3p/BPTF/c-myc axis [325]. In breast cancer, EV circCARM1 originating from cancer stem cells modulates glycolysis in breast cancer cells through the miR-1252-5p/PFKFB2 pathway [326]. Furthermore, miRNA-21a-5p found in MSC-derived EVs has been demonstrated to inhibit the expression of the rate-limiting glycolytic enzyme phosphofructokinase muscle isoform (PFKM) in renal tissues, thereby suppressing glycolysis [327]. Similar inhibitory effects are noted with miR-620 [328], lnc-ABCA12–3 [329], and circFNDC3B [330], which are also delivered by EVs in ESCC.

These findings highlight the critical importance of glucose metabolism in both tumor biology and tumor immunity. Tumor cells facilitate their proliferation and evade immune responses by modifying glucose metabolism within themselves and in adjacent cells, with EVs serving a vital regulatory function in these processes. A deeper understanding of the specific mechanisms underlying glucose metabolism in tumor immunity may contribute to the development of innovative metabolic-targeted therapeutic strategies, ultimately improving the effectiveness of immunotherapies.

Lipid metabolism

It is well established that under nutrient-limited conditions, such as reduced glucose availability, tumor cells can reprogram their lipid metabolism by enhancing fatty acid synthesis, uptake, oxidation, and breakdown. These adaptations provide the energy and structural components necessary for rapid growth and proliferation. Furthermore, lipid metabolic reprogramming contributes to the establishment of an immunosuppressive TME, which facilitates tumor cell survival and invasion, as well as playing a significant role in immune evasion and the development of drug resistance [331]. Such processes are observed across various tumor types and immune-related cells. A prominent feature of the immunosuppressive TME is the polarization of macrophages toward the M2 phenotype, which is characterized by increased lipid synthesis, fatty acid oxidation, and mitochondrial respiration [332]. Notably, this phenotype can be transmitted to recipient cells via EVs derived from the macrophages themselves. In epithelial ovarian cancer (EOC) tissues, EVs from PD-L1⁺ TAMs can upregulate carnitine palmitoyltransferase 1 A (CPT1A) expression in CD8⁺ T cells through the activation of PPARα, thereby promoting fatty acid oxidation. This process results in increased apoptosis and exhaustion of CD8⁺ T cells, ultimately facilitating EOC metastasis [333].

Conversely, EVs secreted by hepatocytes significantly enhance the M1 polarization of Kupffer cells (KCs) by activating the NOX2 and NF-κB signaling pathways, which promotes the production of ROS. This polarization leads to the overproduction of inflammatory cytokines, including TNF-α. The resulting excess of inflammatory factors further increases the expression levels of genes associated with intracellular free fatty acid uptake and lipogenesis, such as fatty acid synthase (FASN) and SREBP-1c, in palmitic acid-treated LO2 cells, while concurrently suppressing the expression of genes related to fatty acid oxidation, such as CPT-1 and PPARα. This dynamic promotes lipid accumulation and hepatocellular injury [334]. These findings suggest that EVs serve as mediators in lipid-related immune regulation.

Upon examining the specific mechanisms underlying these phenotypic changes, a strong association with the cargo carried by EVs was observed. FASN, a key enzyme in the fatty acid synthesis pathway, has been shown to be transferred from EVs derived from various malignancies to recipient cells. This transfer enhances the fatty acid synthesis capacity of these cells, thereby promoting tumor cell proliferation and growth [335–337]. Additionally, CD36, a major lipid transporter with well-documented pro-tumorigenic effects, can also be transferred between cells via EVs [338]. This transfer alters the fatty acid content in recipient cells, consequently promoting tumor progression and the formation of an immunosuppressive microenvironment [339]. EVs derived from pancreatic ductal adenocarcinoma carry the CD44v6/C1QBP complex, which has been implicated in promoting tumor metastasis [135]. CD44, a key marker of cancer stem cells, is closely associated with increased lipid synthesis in tumor cells [340]. Notably, inhibiting CD44 expression has been shown to reduce de novo lipid synthesis in colon cancer cells [341]. Furthermore, CD44 present in EVs from lymph node metastatic tumors has been demonstrated to stimulate fatty acid oxidation in primary tumors by activating Yes-associated protein (YAP) and downstream CPT1A, a fatty acid shuttle enzyme [134]. Collectively, these findings indicate that proteins delivered by EVs play a significant role in lipid regulation within recipient cells.

EVs are capable of transporting enzymes related to fatty acid oxidation. In EVs derived from colorectal cancer cells, HSPC111 plays a critical role in regulating lipid metabolism within CAFs by phosphorylating ACLY, which results in elevated acetyl-CoA levels. This accumulation of acetyl-CoA subsequently promotes increased histone H3K27 acetylation in CAFs, thereby enhancing the expression and secretion of CXCL5. The CXCL5-CXCR2 axis further stimulates EV release, creating a positive feedback loop that fosters a pre-metastatic microenvironment conducive to colorectal cancer progression [342]. Moreover, research indicates that the molecules involved in lipid metabolism differ among various gemcitabine-sensitive pancreatic cancer (PC) cell lines. Notably, acyl-CoA dehydrogenase (ACADM) has been shown to bolster tumor resistance by promoting the consumption of unsaturated fatty acids, which in turn reduces levels of intracellular lipid peroxides and ROS [343].

From the perspective of lipid storage and utilization, EVs play a critical role in the direct transport of lipid molecules, including cholesterol, sphingomyelin (SM), glycosphingolipids, and phosphatidylserine [344], to recipient cells. This transfer can significantly alter the metabolic state of these cells, thereby contributing to the formation of the TME. A lipidomic analysis of colorectal cancer cells has identified 500 distinct types of lipids [345]. Among these, ether lipids or phosphoinositides [88], can be transferred to various recipient cells, including regulatory T cells. This transfer is crucial as it facilitates participation in signaling pathways, enhances immunosuppressive functions, and regulates the metabolic and functional states of immune cells. Moreover, these lipid molecules are essential components of cell membranes. By transferring these lipids, EVs can modify the composition and functionality of the recipient cell membrane, thereby influencing signal transduction and overall cellular behavior.

From a gene regulation perspective, EVs are known to carry non-coding RNAs that play a crucial role in regulating lipid metabolism, thus influencing lipid metabolic pathways in recipient cells through the modulation of gene expression [346]. For instance, miR-15a-5p can inhibit the expression of ACSS2, which leads to a reduction in acetyl-CoA activity, decreased histone H4 acetylation, and inhibited acetate uptake, ultimately suppressing fatty acid synthesis in lung cancer cells [347]. Furthermore, fatty acid transporter protein 2 (FATP2) can be upregulated by circRNA_0013936 present in EVs derived from bladder cancer (BCa). This circRNA acts as a sponge for miR-320a, activating the JAK2 signaling pathway and enhancing expression. This mechanism increases the immunosuppressive activity of PMN-MDSCs, resulting in a significant inhibition of CD8 + T cell function and facilitating immune evasion in bladder cancer [206].

From the perspective of protein modification, N-acetyltransferase 10 (NAT10) is significantly enriched in EVs derived from esophageal squamous cell carcinoma cells. It enhances lipid metabolism by mediating the ac4C modification of FASN, which in turn facilitates the M2 polarization of macrophages [348].

Protein metabolism

Tumor cells undergo metabolic reprogramming to meet the demands of rapid proliferation, a phenomenon that is evident in protein metabolism through increased protein synthesis, altered amino acid metabolism, modifications in protein degradation pathways, and abnormal activity of metabolic enzymes. Notably, malignant cells enhance glutamate release via the cystine/glutamate transporter xCT, which facilitates the cystine-glutamate exchange necessary for maintaining redox homeostasis in cancer cells and promoting disease progression [349]. Furthermore, the aberrant activity of metabolic enzymes, such as pyruvate kinase M2 (PKM2), significantly influences the flux of metabolic pathways, thereby further contributing to the metabolic adaptations of cancer cells.

The TME significantly influences the protein metabolism of surrounding cells by releasing EVs, thereby fostering the development of an immunosuppressive microenvironment. Inhibition of xCT, either through pharmacological agents or gene knockout, markedly elevates PD-L1 levels in EVs derived from melanoma cells, which subsequently promotes macrophage polarization towards the M2 phenotype [350]. Furthermore, HMGB1 mRNA can enhance the expression of the RICTOR protein, which is linked to the aberrant activation of glutamine metabolism in tumor cells. Within the TME, the interaction between HMGB1 and RICTOR may activate glutamine synthetase via the mTORC2-AKT-C-MYC signaling pathway, while concurrently inhibiting glutamate dehydrogenase through SIRT4 in the mTORC1 signaling pathway. This interaction ultimately increases glutamine utilization by tumor cells. Such epigenetic crosstalk can lead to an increased release of PD-L1-positive EVs, diminish the therapeutic efficacy of anti-PD-L1 inhibitors, and further promote tumor immune evasion [351].

EVs are capable of transporting amino acids and their intermediate metabolites. Through experiments utilizing ¹³C isotope labeling, it has been demonstrated that EVs derived from CAFs can directly deliver amino acids to nutrient-deprived cancer cells via a mechanism akin to macropinocytosis. This process concurrently inhibits oxidative phosphorylation in the mitochondria of recipient cells, enhances glycolysis, and promotes glutamine-dependent reductive carboxylation [352].

From the perspective of gene regulation, EVs originating from esophageal cancer cells that carry Circ-SFMBT2 can diminish the proliferation and invasion of EC cells by upregulating the expression of solute carrier family A1 member 5 (SLC1A5) while simultaneously inhibiting glutamine metabolism [353].

These findings underscore the pivotal role of EVs in dysregulated protein metabolism and the establishment of an immunosuppressive microenvironment within tumors, thereby offering new strategies and targets for future cancer therapies.

Energy metabolism

Mitochondria are integral to cellular metabolism and energy production. They facilitate the conversion of various metabolic fuels into the energy required by the cell through several metabolic pathways, including the tricarboxylic acid (TCA) cycle, fatty acid oxidation, the electron transport chain (ETC), and oxidative phosphorylation. These pathways not only generate energy but also supply precursor molecules essential for the synthesis of biomolecules and play a crucial role in regulating nuclear transcription, thereby enabling the cell to adapt to its diverse metabolic requirements [349].

Mitochondrial dysfunction plays a crucial role in the development and maintenance of malignant tumors. For instance, specific mitochondrial proteins, such as signal transducer and activator of transcription 3β (STAT3β), can influence the mitochondrial ETC, thereby impacting cellular apoptotic processes [354]. Dysfunctional alveolar macrophages are characterized by abnormal mitochondrial activity [355]. Mesenchymal stem cell-derived EVs are capable of transferring mitochondrial components to alveolar macrophages in a dose-dependent manner [356]. This transfer can replenish damaged mitochondria, enhance mitochondrial DNA (mtDNA) levels, increase mitochondrial membrane potential (MMP), boost oxidative phosphorylation (OXPHOS) activity, and elevate ATP production, while concurrently reducing mitochondrial reactive oxygen species (mROS) stress in macrophages under LPS challenge. This process ultimately promotes the transformation of macrophages to an anti-inflammatory phenotype [357]. Similarly, alveolar macrophages that uptake EVs derived from alveolar epithelial cells demonstrate increased Ca^2 + reactivity, which regulates mitochondrial biogenesis and mtDNA encoding through the calcineurin (CaN)-PGC-1α pathway, thereby maintaining the M2 phenotype [358].

CD39 catalyzes the hydrolysis of extracellular ATP (eATP) to generate adenosine (ADO), a process that fosters the production of immunosuppressive adenosine. This mechanism inhibits the activity of immune cells, including T cells and NK cells, ultimately contributing to resistance against tumor immunotherapy. EV circTMEM181, derived from HCC, can further enhance the immunosuppressive microenvironment by upregulating CD39 expression, which results in resistance to anti-PD-1 therapy. Research indicates that targeting CD39 on macrophages and inhibiting the ATP-adenosine pathway may restore immune cell function and enhance the efficacy of anti-PD-1 therapy [359]. Additionally, the EV-mediated transfer of key energy metabolism products, such as acetyl-CoA, not only participates in metabolic processes but also functions as a signaling molecule that regulates gene expression and influences epigenetic modifications [348].

In summary, EVs play a crucial role in modulating immune cell function within the TME by regulating various metabolic pathways, including glucose, lipid, amino acid, and energy metabolism. This metabolic reprogramming not only influences tumor cell growth and survival but also contributes to immunosuppressive or immune-activating roles in tumor immunity through diverse mechanisms. Continued research in this domain has the potential to yield significant insights into the mechanisms underlying tumor immune evasion and may pave the way for novel strategies in cancer diagnosis and therapy.

EV-related tumor immunotherapy

The incidence of cancer is steadily increasing and is recognized as a significant threat to human life and well-being. Traditional cancer treatments typically involve radiotherapy, chemotherapy, and surgery. However, in recent years, cancer immunotherapy has emerged as a prominent area of research due to its ability to boost the immune system, its broad applicability across different types of cancers, and its long-lasting effectiveness [360]. Immunotherapy has shown promising anti-tumor effects in various types of cancer, particularly in solid tumors like melanoma, where it has achieved notable success and has become a valuable tool in fighting cancer [361]. This approach to tumor immunotherapy focuses on activating the body’s immune system using natural or laboratory-produced substances. By harnessing the body’s own immune response and coordinating the actions of multiple immune cells, immunotherapy helps the body identify and eliminate cancer cells, thus combating tumors. Compared to traditional cancer treatments, immunotherapy has lower toxicity and fewer side effects on normal tissues, while also targeting both primary and secondary tumors. Additionally, the immune memory developed during immunotherapy provides extended protection for the body, effectively reducing the risk of tumor recurrence and metastasis, and potentially leading to complete tumor eradication. Immunotherapy can be categorized into active immunotherapy (e.g., tumor vaccines), passive immunotherapy (e.g., monoclonal antibodies, cell therapy), and non-specific immunomodulator therapy (e.g., cytokines, immunomodulators, immune checkpoint-blocking ICIs) [6]. These different approaches aim to elicit immune responses through various mechanisms, enhancing the immune system’s ability to target tumors and ultimately improving outcomes and survival rates for cancer patients.

TEXs have been found to carry immunosuppressive molecules like FasL, TGF-β1, TRAIL, PD-L1, and NKG2D ligands, contributing to immunosuppression. Conversely, EVs can also inhibit tumor growth and invasion by creating an immunosuppressive microenvironment, aiding in antigen presentation, and supporting the elimination of tumor cells by various immune cells such as CD4⁺ and CD8⁺ T cells, as well as NK cells [362, 363]. Researchers have proposed the use of EVs in immunotherapy to boost anticancer immune responses and deliver anticancer drugs. In the following sections, we will explore how EVs can impact tumor immunotherapy by influencing their life cycle through treatment and serving as nanocarrier systems.

Targeting EV to inhibit tumor progression and overcome therapeutic resistance

EVs, small extracellular vesicles, can carry harmful disease-related contents like viral miRNAs and proteins. These immunosuppressive factors can enhance tumor metastasis or growth signals that resist therapy. Studies have shown that higher levels of EVs are present in the tumor environment, and injecting mice with excessive EVs leads to pro-tumor effects, such as tumor formation, growth, and metastasis [10–17], angiogenesis [17–21], and metabolic changes in the TME [148]. Additionally, exosome secretion contributes to tumor drug resistance by transporting drugs out of cancer cells, reducing drug effectiveness. Research has demonstrated that pretreating human malignant melanoma with a proton pump inhibitor not only increases cellular uptake of cisplatin but also decreases EV release from tumor cells [364]. This controlled release of EVs can be targeted with therapeutic strategies developed by researchers [365].

Inhibition of EV formation

Oncological treatments and related therapies have the potential to impact tumor cells and immune cells, leading to changes in the secretion of EVs that contain antigens or immunomodulatory molecules, influencing the immune system’s response. It is crucial to note that the majority of TEXs exhibit immunosuppressive properties, underscoring the importance of targeting these EVs to mitigate their harmful effects. A quantitative high-throughput screening (qHTS) assay conducted in prostate cancer cells has identified several compounds such as tipifarnib, sertaconazole, ganbaba, ketoconazole, and triadimefon as inhibitors of EV biogenesis [366]. Conversely, sitafloxacin, capillaries, SB218795, fenoterol, niflumizole, and spray tetrazolium have been validated as activators of EV biogenesis. Researchers have successfully targeted disease-associated harmful EVs using inhibitors of EV biogenesis. For instance, ketoconazole (KTZ), an FDA-approved antifungal drug, has been shown to inhibit EV secretion in sunitinib-resistant clear cell renal cell carcinoma (ccRCC) cell lines, thereby enhancing the effectiveness of sunitinib [367]. Tipifarnib has been found to inhibit the biogenesis and release of EVs by interfering with both ESCRT-dependent and non-dependent pathways, leading to reduced levels of renal cell carcinoma cell surface PD-L1 expression [237].

Chemical inhibitors targeting molecules like sphingomyelinase inhibitors (GW4869) [87] and proton pump inhibitors [364] have been shown to reduce EV secretion in model cells. However, these inhibitors have pleiotropic effects on various cell functions, not just EV secretion, limiting their effectiveness. Understanding the molecular mechanisms involved in vesicle formation and fusion with the plasma membrane in MVB cells could aid in identifying more specific and potent inhibitors of EV secretion. For example, the gene VPS33B plays a crucial role in vesicle transport and membrane fusion in late endosomes and lysosomes [368], with knockdown of VPS33B leading to significant inhibition of EV secretion and delayed onset of leukemia [105]. Additionally, the autophagy-related gene Atg5 regulates late endosome acidification and promotes EV production and tumor metastasis. Knockdown of Atg5 reverses this process [369].

Several studies have also focused on developing fragment peptide drugs targeting EV release. Syntenin, a 32 kDa PDZ tandem (PDZ1 and PDZ2) intracellular protein, plays a crucial role in regulating EV biosynthesis and secretion. It facilitates the transformation of ILVs into MVBs, which then merge with the plasma membrane to release EVs. Knocking down Syntenin has been found to decrease the concentration and secretion of small extracellular vesicles in lung cancer cells [370]. Syntenin interacts with various proteins, including Syndecans, with the help of its PDZ structural domain. Leblanc and colleagues used high-throughput time-resolved fluorescence resonance energy transfer (HTRF) technology to screen a library containing 139 potential PDZ inhibitory fragments, identifying a fragment named C58 that exhibited strong inhibition of the Syntenin-syndecan interaction. Subsequent experimental validation revealed that C58 could significantly impede the migration and colony formation of MCF7 breast cancer cells, as well as alter breast cancer sphere formation and EV release. These findings highlight the promising potential of C58 as a selective inhibitor of EV release [371].

Depletion of EVs in the circulatory system

The impact of immunotherapy on the survival time of EVs in the body is a topic of interest. It is suggested that immunotherapy may influence the metabolism and stability of EVs, potentially prolonging their presence in the body to strengthen the immune response.

Some studies propose the use of external devices like blood purifiers to remove harmful EVs as a potential therapeutic approach. The NCT04453046 clinical trial focuses on removing immunosuppressive EVs in patients with head and neck squamous cell carcinoma (HNSCC) through blood purification [372]. This strategy has shown success in removing viral particles in patients with hepatitis C and HIV, as demonstrated in previous clinical trials such as NCT04595903 which targeted COVID-19 treatment by eliminating the SARS-CoV-2 virus from the bloodstream [373]. In the HNSCC study, patients received pembrolizumab alongside blood purification as a first-line treatment for squamous cell carcinoma of the head and neck. The therapeutic efficacy of this combination therapy was evaluated by observing the dynamics of EV elimination and recovery before, during, and after treatment as a secondary study endpoint [372].

The cytolytic effect of rituximab on tumour cells was significantly enhanced when EVs were removed from B-cell lymphoma patients [225]. This phenomenon may extend beyond rituximab to antibodies targeting other B-cell surface antigens like CD19, CD20, CD22, CD23, CD37, CD40, and HLA-DR [374–376], which are currently being tested in clinical trials. These studies may suggest that targeting EVs could boost the effectiveness of therapeutic antibodies, offering a novel approach to cancer treatment.

Altering the composition of EVs

The composition of EVs is influenced by the specific cell type and condition from which they originate. Sirtuin 1, a NAD⁺-dependent deacetylase, is crucial for maintaining the normal function of the V-ATPase, which is responsible for proper acidification of lysosomes and protein degradation. A reduction in SIRT1 levels can lead to impaired lysosomal function, resulting in fewer degraded MVBs, larger MVBs, and altered EV secretome. This alteration can potentially enhance breast cancer cell proliferation and invasion [377].

Some oncological treatments, particularly immunotherapy, can modify the immune phenotype of the originating cell, consequently impacting the composition of EVs. This alteration further influences the tumor immune microenvironment. Numerous studies have demonstrated that EVs derived from tumors hinder the proliferation and cytotoxicity of CD8⁺ T cells, thereby facilitating immune evasion by the tumor. Notably, the removal of EVs containing PD-L1 inhibits tumor growth, even in anti-PD-L1 antibody-resistant models [29]. The triazine compound 6J1 inhibits the endosomal transport of PD-L1 by activating Rab5, resulting in the accumulation of PD-L1 within endocytosed vesicles. Concurrently, it enhances the secretion of extracellular vesicular PD-L1 through the activation of Rab27, which leads to a decrease in PD-L1 membrane levels on tumor cells, thereby significantly enhancing the anticancer immune response [378]. Intercellular adhesion molecule (ICAM) plays a crucial role in immune surveillance and inflammatory responses by primarily promoting cell adhesion. The interaction between ICAM and LFA is essential for T cell activation, immune cell migration to inflammatory sites, and immune cell interactions during antigen presentation. Guo et al. discovered that the adhesion molecule ICAM-1 co-localizes with PD-L1 on EVs. Blocking ICAM-1 prevented TEX from adhering to T cells via ICAM-1-LFA-1 interactions, resulting in a marked decrease in the interaction between PD-L1 on TEX surface and PD-1 on T cells [291]. Similarly, knockdown of histone lysine-specific demethylase 1 (LSD1) [379] and overexpression of miR-16-5p [380] reduced the exocytosis of PD-L1 and restored T cell responses.

Other immune-related therapeutic approaches that directly modify EV contents are considered cutting-edge strategies focused on enhancing immune activation or modulating immune responses. This can be achieved through loading EVs with antigenic, immunomodulatory molecules, or pharmacological agents, as will be further discussed.

Enhanced immune cell activation through EV-mediated antigen presentation and cytokine production

EVs can express ligands on their surface that bind to receptors highly expressed on tumor cells or immune cells, leading to specific functions. For instance, NKG2D ligands on dendritic cell-derived EVs interact with NKG2D receptors on NK cells, activating NK cells and promoting proliferation [381]. Conversely, EVs with NKG2D ligands can inhibit NKG2D-mediated cytotoxic responses [382]. Mature DCs stimulated by LPS release EVs containing high levels of MHC-1 and ICAM-1, which are taken up by other DCs through LFA-1 and ICAM-1 interaction, facilitating the transfer of processed antigens and peptide-MHC complexes to target cells [383]. While these DCs cannot reprocess antigens, they can enhance T cell activation by presenting EV-associated MHC peptide complexes and exogenous antigens [383]. Furthermore, dendritic cell-derived EVs can transport TLR ligands to other DCs, activating them and boosting antigen presentation and inflammatory cytokine production [384]. Activated immune cells are more likely to take up EVs, gaining information on tumor antigens or immunomodulatory molecules. Activated T cells can attract EVs secreted by dendritic cells through LFA-1 receptors [291], indicating an increased ability of activated T cells to interact with and internalize EVs.

Immunotherapy can impact the life cycle of EVs in various ways, ultimately boosting the immune system’s ability to target tumors. These treatment approaches necessitate thorough research and clinical trials to assess their effectiveness and safety, improving their utilization in tumor therapy. As the field of EV research continues to develop, a deeper understanding of its role in immunotherapy will be gained.

EV-Based nanocarrier systems for therapeutic applications

EVs, a naturally occurring nanoscale carrier system, are increasingly recognized as a crucial element in medical therapy (Table 5). These small vesicles, naturally released by cells, possess a remarkable capacity to transport drugs, genes, and other therapeutic molecules safely and effectively within the body. Their compatibility with biological systems and minimal likelihood of causing immune responses make them well-suited for drug delivery. EVs can target specific diseased cells by recognizing surface markers, leading to higher drug concentrations at the site of disease while minimizing harm to healthy tissues. Moreover, the lipid and protein composition of EVs enables them to traverse biological barriers, such as the blood-brain barrier, facilitating access to challenging anatomical locations [385]. Consequently, EVs exhibit significant promise as nanocarriers for future drug delivery and therapeutic interventions.

Table 5.

Application of EVs as drugs or drug carriers in immunotherapy

EV Source	Modification Method	Target	Biofunctions	Model	Ref.
γδ-T	N/A	T cell	Promoting T cell activity, antiviral and antitumor function		[386]
	N/A	EBV-associated tumor cells	Killing EBV-associated tumor cells and promoting EBV antigen-specific CD4+ and CD8+ T cell expansion		[387]
M1 macrophage	OX40Loverexpresses lentivirus	macrophage, T cell	Reprogramming M2-like macrophages into M1-like macrophages and promote the activation and proliferation of CD8+ T cells	Breast cancer mouse	[388]
DC	Functional domain of HCC targeting peptide (P47-P), alpha-fetoprotein epitope (AFP212-A2), and high mobility Group nucleosome binding protein 1 (N1ND)	DC	DC recruitment, accumulation and activation	In situ HCC tumor mice	[389]
	GE11 peptide	T cell	Increasing activity of CD8+IFN-γ+ T lymphocytes	Breast cancer mouse	[236]
	EGFR and CD 3 antibody	T cell	Activating T cells to attack cancer cells that express EGFR		[390]
M1 macrophages	Cancer cell nucleus	T cell	Increasing T cell production and regulating immunosuppressive TME		[391]

It is essential to acknowledge that the diversity of EVs, characterized by variations in size, content, and biogenesis, presents both opportunities and challenges. This heterogeneity enables exosomes to target various cell types and tissues, which is particularly advantageous in personalized medicine, where customized therapies are vital. Conversely, the variability in exosome characteristics complicates their application, particularly concerning standardization, quality control, and reproducibility [392]. The heterogeneity of exosomes implies that individual vesicles do not share identical chemical or physical properties. Notably, even exosomes derived from a single cell can exhibit compositional heterogeneity, indicating the presence of functional subpopulations, irrespective of the homogeneity of the source [393]. Consequently, the complexity of exosome-based drug delivery systems hampers the development of universal targeting strategies, as different types of exosomes may interact with target cells in unique ways. Variations in exosome composition can influence their capacity to effectively deliver cargo, resulting in inconsistent therapeutic outcomes. Addressing this variability necessitates advancements in methods for exosome isolation and characterization to ensure reliable and consistent drug delivery. The CP05 peptide anchoring modification can maintain the natural size and morphological characteristics of exosomes, facilitating the capture of exosomes from diverse sources by binding to the exosome surface protein CD63. Additionally, CP05 peptides can function as a bridge, conjugating target molecules to modify exosomes without compromising their in vivo distribution [394]. Similarly, aptamers—short oligonucleotides that can bind multiple ligands—represent an emerging strategy for exosome surface functionalization. By linking aptamers to the membrane surface molecules of exosomes, this approach enhances exosome targeting by enabling interactions with specific ligands [395].

Quality assessment of EV carrier systems

Due to their excellent stability, biocompatibility, low immunogenicity, and unique surface protein expression derived from their parental cells, EVs are widely utilized as efficient drug-delivery vehicles. As a promising clinical therapeutic option, it is crucial to establish a comprehensive quality assessment system for EVs and their derivatives, encompassing isolation, purification, quantification, characterization, drug metabolism, biosafety, toxicity, and potential side effects, prior to their clinical use.

Current studies have identified six strategies for EV separation: ultracentrifugation, ultrafiltration, size exclusion chromatography, precipitation, immunoaffinity-based capture, and microfluidics [9]. Each method presents unique advantages and challenges that require attention. Ultracentrifugation offers advantages such as affordability, large sample volume capacity, and the ability to separate high concentrations of EVs [396]. However, it is a time-consuming process that poses a risk of EV damage due to high-speed centrifugation and requires complex equipment. A further advancement, known as differential centrifugation, involves multiple rounds of centrifugation to isolate target EVs from cellular debris, larger vesicles, and proteins. The process of EV isolation involves frequent user intervention to remove supernatant and precipitate, as well as establishing a spin cycle. There may be loss of EVs during the repeated removal of supernatant and transfer of samples between tubes. Despite these drawbacks, the method of differential centrifugation has been extensively tested and proven to yield moderate amounts of EVs with reasonable purity. Another method, ultrafiltration, separates EVs based on size differences from other particles. While this strategy is fast and portable, it has downsides such as low purity, shear stress, EV loss, and clogging. Size exclusion chromatography also relies on size differences to accurately separate EVs without damaging them, but it may be more time-consuming and require further development. The precipitation method alters the solubility of EVs and offers advantages such as ease of handling, suitability for large sample sizes, and minimal damage to EVs [397]. However, this method can be time-consuming and may precipitate other particles like polymeric materials and proteins. Immunoaffinity-based capture is another technique that relies on the interaction between antibodies and antigens [398]. This method offers high purity and the possibility of subtype analysis, but it comes with drawbacks such as high cost, low yield, risk of antigen blocking, and potential loss of EV function. The microfluidic strategy for EV isolation is characterized by its affordability, time efficiency, and minimal sample requirement, but it may have lower sensitivity compared to other methods [399].

Previous studies have identified several proteins commonly used as markers for EVs, including CD63, CD81, CD9, MHC class II or class I molecules, heat shock proteins HSC70/HSP73 and HSP70/HSP72, flotillins, and actin [44]. However, these proteins are not exclusive to EVs and are found in various EVs. Therefore, they cannot serve as specific markers for EVs or small EVs on their own. It is important to explore more specific methods for isolation and identification. CD63 is often enriched in late endosomes [99], making it a preferred marker for isolating endosome-derived EVs. Nonetheless, some studies have reported the absence of CD63 in small EVs secreted by certain cell types [59], necessitating the use of alternative markers such as CD81 or CD9. Additionally, since CD63 is also present in large EVs, it is crucial to first eliminate these large EVs before using CD63 as a specific marker for EV isolation [400]. These findings underscore the importance of employing multiple isolation and identification techniques in small EV studies to accommodate the diverse characteristics of different cell types and EV subtypes.

Studying the mechanisms by which EVs work reveals a close relationship between EV dose and their effects. EVs from human glioblastoma have been shown to stimulate human brain microvascular endothelial cells, promoting tumor proliferation, motility, and tube formation in a dose-responsive manner [17]. The debate and research surrounding EV dosing have highlighted the current limitations in accurately assessing EVs at the level of individual vesicles. Therefore, the development of standardized methods for EV administration is crucial. To enhance the precision of EV administration, it is recommended to employ multiple methods for evaluating EV dose synergistically. Various approaches, such as cell equivalents, protein concentrations, and specific quantitative analytical measurements using tools, each have their own advantages and disadvantages. For instance, while the protein method offers a quick and cost-effective way to assess total protein levels, it may also measure proteins unrelated to EVs and fail to indicate biologically active components. Other methods for evaluating extracellular vesicular dosing include TRPS, NTA, ELISA, cell equivalents, and flow cytometry.

A critical consideration in the use of EVs in clinical trials is adherence to good manufacturing practice (GMP). EVs employed in these trials must conform to GMP standards, which encompass three essential components: upstream cell culture, downstream purification processes, and quality control of the EVs. These aspects are closely linked to the safety, stability, and potential side effects of EVs in clinical applications. The primary safety concern regarding EVs as potential drug delivery systems is their immunogenicity. EVs may carry tumor or pathogen-derived peptides, which can be presented via major histocompatibility complex (MHC) receptors, thereby initiating interactions with components of the immune system. While this characteristic is exploited in cancer immunotherapy [392], it may also result in adverse immune responses, including allergic reactions and cytokine release syndrome [401].

In addition to immunogenicity, the stability of EVs is another critical issue for their clinical applications. EVs, due to their endogenous biosynthetic mechanisms, are considered to be highly stable in vivo [392]. However, various factors, such as storage conditions, delivery methods, and in vivo circulation, can affect their stability. Modifying the surface characteristics or composition of EVs is considered a strategy to enhance their stability, ensuring that they remain intact and functional during transport to target cells. However, it has also been reported that EVs have a limited half-life, ranging from 2 to 20 min, which is much shorter than that of liposomes, whose half-life can extend to several days [392]. PEGylation (polyethylene glycol modification) can significantly extend the circulation time of EVs, but it may also cause accelerated clearance, as IgM antibodies may be generated against the PEG modification during repeated administration [392]. Furthermore, the side effects associated with EV-based therapies should not be overlooked. Off-target delivery or unintended organ accumulation of EVs can lead to toxicity and adverse reactions [401]. To address these issues, researchers are investigating targeted EVs, which can be designed to specifically direct the vesicles to the desired tissues or cells, thus reducing the risk of systemic toxicity and improving therapeutic efficacy. Potential tumor-promoting activities should also be considered, especially for stem cell-derived EVs with angiogenic activity, which may have tumor-promoting effects in existing tumors. Additionally, CD47 modification strategies have significant side effects. For example, CD47-modified EVs mimic the natural process by which red blood cells and platelets avoid phagocytosis by macrophages/monocytes, which can lead to a reduction in circulating red blood cells [402].

In conclusion, while EVs hold significant promise as therapeutic delivery systems, it is imperative to thoroughly assess their safety, stability, and potential side effects. Key strategies aimed at mitigating risks, including enhancing targeting capabilities and improving stability, will be crucial for the successful translation of EV-based therapies into clinical practice.

Extracellular vesicular self-realisation immunotherapy

EVs, as naturally occurring nanoscale bioparticles, have demonstrated significant potential in immunotherapy by effectively activating the immune system. For example, EVs derived from tumor cells can transport tumor antigens and be absorbed by immune cells like DCs, initiating an acquired immune response and guiding immune cells to precisely target tumor cells. Moreover, EVs can carry immunomodulatory molecules, including immune checkpoints, directly to immune cells, effectively counteracting the immune suppression mechanisms of tumors and their microenvironment, thereby boosting the immune system’s ability to combat tumors. Depending on their cell of origin, EVs exhibit specific organ tropisms; for instance, dendritic cell-derived EVs tend to accumulate in the spleen. Conversely, endothelial EVs from the brain have a tendency to accumulate in brain tissue [403]. These evidences indicate that integrating EV technology into tumor treatment strategies not only enhances immune cells’ capacity to identify and attack tumors but also helps regulate the overall immune response, thereby amplifying the therapeutic outcomes.

Numerous studies have identified that DC-produced EVs express a wide array of tumor markers, including MHC-I, co-stimulatory molecules, and heat shock proteins (HSPs), crucial for antigen presentation and T cell activation. These EVs have shown in vitro to elicit CD8⁺ T cell-mediated antitumor responses [391, 404, 405]. For instance, the clinical trial NCT01159288 utilized dendritic cell-derived EVs carrying antigen-presenting MHC class I/II and co-stimulatory molecules as immunotherapeutic agents for non-small cell lung cancer patients post-chemotherapy. While EV therapy enhanced NK cell antitumor activity, it did not elicit a significant adaptive immune response [406]. In another clinical study involving patients with MAGE3 + advanced malignant melanoma (MM) treated with dendritic cell-derived EVs, disease stabilization was observed; however, no response from MAGE-specific CD4⁺ and CD8⁺ T cells in peripheral blood or delayed-type hypersensitivity (DTH) response was detected [407]. Despite the efficacy of dendritic cell-derived EVs in these clinical trials, specific T-cell responses were not evident. To address this, EVs derived from Toll-like receptor 4 ligand (TLR4L) or IFNγ matured DCs were used in a new phase II study to enhance the T cell response induced by dendritic cell-derived EVs. Unfortunately, the results showed that while these EVs enhanced NKp30-dependent NK cell function, they did not elicit an antigen-specific T cell response [408].

γδ-T cells, also known as γδ-T-EVs, are a type of innate T cells that can exhibit lytic activity independently of MHC dependence [409]. Multiple studies have shown that γδ-T-EVs have the ability to directly eliminate stressed cells, display strong pro-T cell activity, and possess potent antiviral and antitumor functions [386]. On the other hand, Vδ2-T cells, which originate from phosphoantigen expansion, release EVs containing death-inducing ligands such as FasL and TRAIL, activation receptors for NK cells like NKG2D, immunostimulatory ligands including CD80 and CD86, and antigen-presenting molecules like MHC class I and class II. Vδ2-T-EVs are capable of targeting and effectively killing EBV-associated tumor cells through the FasL and TRAIL pathways, while also promoting the expansion of EBV antigen-specific CD4⁺ and CD8⁺ T cells [387].

Studies have indicated that EVs produced by cells associated with disease or infection can enhance immune responses. For instance, EVs derived from A549 lung cancer cells exhibit altered cargo composition following respiratory syncytial virus infection, leading to the activation of the innate immune response through the induction of cytokines and chemokines in human monocytes and airway epithelial cells [240]. Additionally, circulating EVs containing circZNF451 are elevated in LUAD patients with progressive disease compared to those in partial remission post PD1 blockade therapy. These EVs play a role in reshaping the TME by promoting an anti-inflammatory macrophage phenotype and reducing CD8⁺ T cells through the FXR1-ELF4-IRF4 axis [410]. These findings underscore the potential of EVs in innovative therapies aimed at addressing various diseases and injuries.

EVs as natural nanocarriers for drug delivery in Immunotherapy

EVs have shown great potential in transferring biologically active substances between cells, making them valuable as drug carriers in disease treatment. They can not only serve as direct therapeutic agents but also transport therapeutic molecules like siRNAs, miRNAs, and chemotherapeutic agents to target cancer cells and boost anti-cancer immune responses. Compared to other carriers, EVs exhibit low toxicity and immunogenicity, ensuring a safer profile, and demonstrate high stability in the bloodstream due to their cell-derived surface, resulting in a prolonged blood half-life by evading capture by the reticuloendothelial system [411]. These characteristics position EVs as promising drug carriers that have garnered significant attention. An important consideration in EV drug delivery is determining the mode of drug loading, with various methods available such as in vitro loading of EVs from parent cells, loading drugs into parent cells for subsequent release in EVs, and transfecting/infecting parent cells with DNA encoding therapeutic compounds for release in EVs [412]. Each loading method presents distinct advantages and limitations, and the choice can depend on the specific therapeutic drug, disease location, and conditions suitable for the particular EV-encapsulated drug [413].

The selection of the type of drug to be encapsulated within EVs is a critical consideration, taking into account factors such as solubility, stability, molecular size, and interactions. These properties influence the method of drug loading and the optimal amount of drug to be loaded. Parameters like the rate of drug release are assessed through ex vivo and in vivo release experiments [414, 415]. Additionally, the route of administration must be carefully chosen based on the intended purpose of drug delivery. Throughout the process of drug loading and release, it is essential to prioritize the stability of EVs, the protective effects of the drug, and any potential toxic side effects on target cells.

EVs have the ability to transport various therapeutic drugs. In the process of loading EVs obtained from parent cells in vitro, lipophilic small molecules can typically be passively loaded into EVs or EV-like vesicles through co-incubation or sonication. For instance, antioxidants, curcumin, anticancer drugs like adriamycin (Dox) [403] and paclitaxel (PTX) [416] can be loaded into EVs or EV-like vesicles through room temperature incubation. On the other hand, hydrophilic molecules such as dopamine may require incubation in a saturated solution to enhance loading [417]. Saponification, compared to passive methods, can lead to an 11-fold increase in loading of large or hydrophilic compounds, provided that the impact of detergents on EV integrity is addressed [418]. Researchers have also discovered that loading parent cells with a drug can lead to its release in the corresponding EV [419]. Additionally, electroporation of purified EVs has been proposed to load exogenous RNA [388], with some studies demonstrating successful electroporation of siRNA into EVs derived from DCs [420]. This method has also been utilized to deliver miRNAs targeting the EGFR [421] or doxorubicin [422] into EVs for breast cancer therapy. Other techniques such as extrusion, freeze-thaw cycles, and chemical transfection have been employed to transfer drugs or active molecules into EVs [423, 424]. EVs have the ability to naturally deliver various genetic materials such as mRNA, miRNA, noncoding RNAs, mitochondrial DNA, and genomic DNA, making them a promising candidate for gene therapy. The process involves introducing DNA encoding therapeutic compounds into parental cells, which are then released in EVs [388]. Lin et al. demonstrated this by inducing M1-type macrophages in mice through infection with OX40L overexpressing lentivirus, leading to the extraction of OX40L M1-EVs [425].

Targeted delivery of immunotherapy based on EV membrane structure realization

EVs have emerged as promising vectors for tumor immunotherapy due to their unique characteristics, including efficient tumor targeting and delivery, ease of customization, immunomodulatory properties, and minimal side effects [426]. On the other hand, challenges in the TME hinder the effectiveness of nanomedicine-based immunotherapy. Factors such as phagocytosis by the reticuloendothelial system (RES) and complement system activation limit the circulation time of nanomedicines [427]. Moreover, obstacles like abnormal vascular structure, dense extracellular matrix, and high interstitial fluid pressure around tumors impede the distribution and penetration of nanomedicines into the tumor tissue [428, 429]. Additionally, elevated levels of GSH and ROS in cancer cells, along with overexpression of P-glycoprotein in cancer stem cells (CSCs), reduce the intracellular uptake and therapeutic potential of nanomedicines [130]. Overcoming these barriers is crucial for the development of more effective nanomedicines in cancer treatment.

EVs offer a significant advantage as drug carriers due to their ability to transport drugs and therapeutic genes, as well as their customizable surfaces through bioengineering with antibodies, fluorescent dyes, and peptides. This customization allows them to meet the demand for both small molecules and large-scale active biologics, optimizing their functionality. Compared to polymer-based carriers, EVs exhibit superior biocompatibility, higher payload capacity, and lower immunogenicity, making them promising drug-delivery vehicles [430]. Their active targeting and specificity enable them to overcome the limitations of traditional drug delivery methods and effectively deliver payloads to target cells [431]. Yong emphasized that for EVs-based nanomedicines to achieve optimal anticancer activity post systemic administration, they must possess five key characteristics: prolonged circulation, enhanced tumor accumulation, deep tumor penetration, effective cellular internalization, and controlled drug release [432].

In order to address these issues, researchers have made various attempts to modify EVs in order to decrease their clearance by the mononuclear phagocyte system (MPS) when used as drug-carrying nanocarriers. These modifications aim to enhance their interaction with target cells, improve their specific distribution in diseased tissues, and minimize their recognition by the immune system.

EVs can be modified with cell-penetrating peptides (CPPs) to aid in their entry into cells. CPPs, also referred to as protein-transducing structural domains (PTDs), are short peptides that facilitate the transport of macromolecules into cells [433, 434]. These CPPs are typically categorized into three main groups: cationic CPPs, amphiphilic CPPs, and hydrophobic CPPs [435]. Structurally, CPPs can be classified into two groups: those rich in arginine and those with amphipathic properties [436]. EVs modified with arginine-rich CPPs can improve internalization by promoting microcellular drinking [437].

However, the application of CPPs in vivo has limitations, including short half-life, endosomal escape issues, lack of specificity for cells and tissues, and potential toxic effects [438, 439]. Researchers have explored various strategies to address these challenges, such as incorporating unnatural amino acids to enhance enzymatic stability and developing activatable or antagonistic CPPs for improved functionality [440]. For instance, SPA, a derivative of substance P, has demonstrated strong antitumor effects as an antagonist. Zhang et al. theorized that SPA, due to its amino acid composition, could potentially function similarly to CPPs in translocating across cell membranes [441]. Additionally, Rao’s team engineered mesenchymal stem cell-derived EVs conjugated with superparamagnetic iron oxide nanoparticles (SPIONs) and anchored with CPP and TNF-α. This modified EV was shown to target cancer cells with heterogeneous characteristics under an external magnetic field in a mouse model of melanoma [442]. These advancements highlight the potential of CPPs in improving targeted drug delivery and therapeutic efficacy in cancer treatment.

Ligands on the surface of EVs can be modified to have high affinity for receptors that are highly expressed on the surface of tumor cells, thereby improving the ability of EVs to bind to tumor cells. Li et al. discovered that the cyclic peptide LXY30 exhibits high affinity and specificity for tumors, effectively detecting α3β1 integrin-expressing NSCLC cells and EVs of tumor origin [443]. Tian et al. demonstrated that immature DCs can secrete EVs expressing RGD, a peptide that specifically binds to αvβ3 integrins highly expressed on tumor surfaces, and carrying the chemotherapeutic drug adriamycin inside the EVs. The modified EVs successfully reached the tumor site after intravenous injection, resulting in significant therapeutic effects [444]. Moreover, a tumor-targeting iRGD-EVs-rMETase was developed by introducing recombinant methioninase (rMETase) into EVs secreted from immature DCs with externally modified RGDs through electroporation. This targeted drug delivery system showed good tumor-targeting function and inhibitory effects [445]. Additionally, the attachment of RGD and magnetic nanoparticles (MNP) to EV in pancreatic cancer was also found to enable targeted delivery of paclitaxel [446].

In addition to RGD, tumor-specific targeting peptides can also be utilized to modify EVs for tumor targeting. For example, the DEX vaccine co-modified with HCC-targeting peptide (P47-P), alpha-fetoprotein epitope (AFP212-A2), and functional domain of high mobility group nucleosome-binding protein 1 (N1ND)-N can facilitate DC recruitment, accumulation, and activation of DCs from HCC tumor-bearing mice in situ [389]. Many epithelial-origin tumors show increased EGFR expression, making EGFR a potential receptor target in cancer drug delivery systems. The GE11 peptide (amino acid sequence YHWYGYTPQNVI) specifically binds to EGFR, enabling surface-modified GE11 peptide EVs to effectively target receptor cells with high EGFR levels [447]. CDCP1, also known as CD318, is a transmembrane protein that is often overexpressed in various tumors [448]. Proteomics and bioinformatic analysis have revealed IR-induced alterations of cargoes in H22 and 4T1 sEVs, with CDCP1 being one of them. Peptides designed and synthesized based on optimal MHC-I binding and immunogenicity for CDCP1 could enhance active CD8⁺-IFNγ⁺ T lymphocytes and suppress the growth of 4T1 tumors in mice [236].

In the realm of EV modification for therapeutic purposes, enhancing the antitumor activity of EVs often involves modifying tumor-associated antigens on their surfaces. An ideal TAA should possess three key characteristics: a pattern of overexpression in tumors, oncogenic properties, and immunogenicity [236]. Researchers have developed EVs loaded with specific MAGE and other peptides, which have shown to partially boost the activity of T-cell nuclear NK cells and demonstrate positive efficacy in certain patients with advanced NSCLC [406]. Another innovative therapy involves synthetic multivalent antibody-targeted EVs (SMART-EVs), which can target both tumor-associated human EGFR and CD3 receptors on T cells. SMART-EVs are engineered to redirect and activate T cells to combat EGFR-expressing cancer cells, displaying potent and specific antitumor effects [390].

In recent years, due to the intricate nature of the TME, targeting a specific type of immune-promoting cell for proliferation, activation, or infiltration alone has proven to be a challenging task in meeting the demands of effective tumor treatment. To address this issue, the concept of a hybrid cell has been introduced. Ma et al. devised a macrophage-tumor chimeric EV that serves the dual purpose of homing in on tumors and activating the immune system. This unique EV was created by incorporating M1 macrophages into the core of cancer cells. Through experiments conducted on mice, it was demonstrated that this EV exhibited strong anti-tumor effects by enhancing T cell production and modifying the immunosuppressive TME [391].

Previous studies have shown that utilizing viral vectors can effectively deliver siRNA with high specificity and efficiency. Building upon this research, the modification of EVs using viral proteins has been proposed as a potential method for enhancing EV delivery of genetic material. The integration of specific viral proteins has been found to greatly improve the efficiency of EV-mediated genetic information delivery, as viral envelope proteins possess strong binding and entry capabilities [449]. It has been observed that EVs derived from EBV-infected cells naturally carry the viral glycoprotein gp350 on their surface. EVs released from HEK293 cells that overexpress EBV gp350 have demonstrated precise localization to CD21-positive B cells [450]. However, caution should be exercised before utilizing these EVs in vivo, as CD21 is also present in other cell types, such as follicular dendritic cells.

Conclusion and prospective

Cancer compromises the immune system by impairing its function and disrupting the production of essential blood cells required for fighting infections [451]. Various treatment modalities, including chemotherapy, targeted therapies, radiotherapy, and high-dose steroids, can induce temporary immunosuppression [452–454]. This highlights the importance of comprehensively assessing the immune status of cancer patients to elucidate the diverse immune profiles and their relationships with tumor origin, stages of progression, and patient demographics. Such insights are vital for informing future therapeutic strategies and mechanistic studies.

The present review outlines the intricate interplay between EVs and tumor immunity [455], highlighting the potential for leveraging this knowledge to develop personalized therapies [456]. EVs, which are nanoscale vesicles secreted by cells, play a dual role in tumor immune regulation [455]. They facilitate immune evasion by tumor cells, thereby promoting tumor growth and metastasis, while also possessing the capacity to carry and deliver immune-regulating factors that activate and modulate the immune response, thereby bolstering the body’s defense against malignant diseases [457]. Despite the extensive documentation of the regulatory functions of EVs, their specific mechanisms warrant further exploration. Key questions remain regarding how EVs selectively package and transport their molecular contents, as well as how these processes are influenced by cellular states and microenvironmental factors. A more comprehensive understanding of the biological characteristics of EVs and their specific roles in tumor immunity could pave the way for innovative therapeutic strategies, ultimately enhancing the specificity and efficacy of cancer treatments.

Recognizing EV-related tumor immunotherapy as a promising emerging strategy [458], we can leverage the unique properties of EVs to enhance tumor recognition and facilitate immune system attacks. EV-based drug delivery also presents new therapeutic avenues for patients. Furthermore, EVs that carry tumor antigens, immune suppressive and activating markers, as well as immune cell chemokines, play critical roles in elucidating patients’ immune status, therapeutic responses, and predicting outcomes of immunotherapy, as well as overall patient prognosis (Fig. 6) [459]. This information provides healthcare professionals with valuable insights into immune status and tumor immune responses, thereby aiding in the selection of optimal treatment strategies and proactively determining therapeutic responses and prognoses. Consequently, this approach enhances the effectiveness of cancer treatment and enables personalized therapy. These findings underscore the dual utility of EVs as both biomarkers and therapeutic tools. However, to fully realize these benefits, it is essential to advance our understanding of EV biology and mechanisms, thereby achieving more precise in vivo distribution, targeting, and release.

Fig. 6.

The clinical potential of EVs. EVs are commonly utilized as biomarkers or therapeutic targets in clinical settings. The concentration of EVs and their cargoes, including nucleic acids and proteins, have the potential to serve as diagnostic biomarkers for tumors and prognostic indicators for treatment. Cancer EVs that hinder the biogenesis, release, or uptake of EVs and eliminate circulating EVs could represent novel targets for anticancer therapy. The low biocompatibility and immunogenicity of EVs make them well-suited for drug delivery systems. EVs have the ability to selectively target diseased cells based on specific surface markers and achieve targeted drug delivery through internal drug delivery and external modification

Current conclusions regarding the role of EVs in intercellular communication predominantly stem from in vitro experiments, which may not accurately reflect physiological conditions due to the concentrations of EVs utilized. The development of mice with specific and complete deficiencies in EVs could help address these limitations. As a result, it is anticipated that in vivo studies exploring the functional roles of EVs in cancer will become more common. Furthermore, additional clinical trials are essential to rigorously validate the efficacy and safety of these strategies (Table 6). Although preclinical models have demonstrated promising therapeutic effects, the clinical application of EVs encounters several challenges, including scalable production, effective drug loading and release, safety assessments, and regulatory approvals [460]. As research advances, EVs are expected to assume an increasingly significant role in cancer immunotherapy.

Table 6.

EVs and their derived drugs targeting tumor immunity in ongoing clinical trials

EV Source	Modification Method	Target	Condition	Phase	NCT number	Current Status
DC	MAGE	T cell ana NK cell	NSCLC	I	NCT01159288	Completed
UCMSC	N/A	myelosuppression	chemotherapeutic acute myeloid leukemia	I	NCT06245746	Not yet recruiting

In conclusion, research on EVs enhances our understanding of tumor immunobiology and holds promise for the development of novel therapeutic strategies, particularly within the realms of precision and personalized medicine. As natural nanoscale carriers, EVs present unique advantages for immune modulation and information transfer, thereby offering innovative perspectives and methodologies for cancer treatment. Investigations into EVs have elucidated their essential roles in tumor development and have indicated new therapeutic avenues that leverage these mechanisms. Future research must confront existing scientific and technical challenges to fully harness the potential applications of EVs in cancer therapy.

Abbreviations

ACADM

Acyl-CoA dehydrogenase

ADO

Adenosine

AFP

Alpha-fetoprotein

APCs

Antigen-presenting cells

ASO

Antisense oligonucleotides

Bca

Bladder cancer

CAR

Chimeric antigen receptor

ccRCC

Clear cell renal cell carcinoma

CPPs

Cell-penetrating peptides

CPT1A

Carnitine palmitoyltransferase 1A

CSCs

Cancer stem cells

CTL

Cytotoxic T lymphocyte

DC

Dendritic cell

DEX

DC-derived exosomes

DPP4

Dipeptidyl peptidase 4

dsDNA

Double-stranded DNA

DTH

Delayed-type hypersensitivity

EAE

Encephalomyelitis autoimmune disease

eATP

Extracellular ATP

ECM

Extracellular matrix

EGFR

Epidermal growth factor receptor

EOC

Epithelial ovarian cancer

ESCC

Esophageal squamous cell carcinoma

EVs

Extracellular vesicles

FasL

Fas ligand

FASN

Fatty acid synthase

FATP2

Fatty acid transporter protein 2

FGF11

Fibroblast growth factor 11

Foxp3

Forkhead box P3

Gal-9

Galactoglucan lectin 9

GMP

Good manufacturing practice

GZB

Granzyme B

HCC

Hepatocellular carcinoma

HLA

Human leukocyte antigen molecules

HMBOX1

Homology box inclusion 1

HNSCC

Neck squamous cell carcinoma

HSPs

Heat shock proteins

ICAM

Intercellular adhesion molecule

IEXs

Immune cell-derived exosomes

IFN-I

Type I interferon

IFN-γ

Interferon-gamma

IL-35

Interleukin-35

IL-6

Interleukin-6

KCs

Kupffer cells

KTZ

Ketoconazole

LEC

Lymphatic endothelial cells

LNs

Lymph nodes

LPS

Lipopolysaccharide

LSD1

Lysine-specific demethylase 1

MDSC

Myeloid-derived suppressor cells

MHC

Major histocompatibility complex

MMP

Mitochondrial membrane potential

MMP

Malignant melanoma

MNP

Magnetic nanoparticles

MPs

Mononuclear phagocyte system

MSC

Mesenchymal stem cells

mtDNA

Mitochondrial DNA

MVB

Multivesicular bodies

NAT10

N-acetyltransferase 10

NB

Neuroblastoma

NK

Natural killer

NPC

Nasopharyngeal carcinoma

NSCLC

Non-small cell lung cancer

NSCs

Neural stem cells

OGT

O-GlcNAc transferase

ONT

Oligonucleotide therapy

OXPHOS

Oxidative phosphorylation

PC

Pancreatic cancer

PD1

Programmed cell death protein 1

PD-L1

Programmed cell death ligand 1

PDCD4

Programmed cell death factor 4

PFK

Phosphofructokinase

PFKM

Phosphofructokinase muscle isoform

PKM2

pyruvate kinase M2

PTDs

protein-transducing structural domains

qHTS

Quantitative high-throughput screening

RES

Reticuloendothelial system

rMETase

Recombinant methioninase

ROS

Reactive oxygen species

sEV

Small extracellular vesicle

SLC1A5

Solute carrier family A1 member 5

SM

Sphingomyelin

SPIONS

Superparamagnetic iron oxide nanoparticles

SS

Sunitinib-sensitive

TAMs

Tumor-associated macrophages

Tc

Cytotoxic T cells

TCA

Tricarboxylic acid

TCR

T-cell receptor

Teff

Effector T cells

TEXs

Tumor-derived exosomes

Th

Helper T cells

TIB

Tumor immune barrier

TLR4L

Toll-like receptor 4 ligand

TLR8

Toll-like receptor 8

TME

Tumor microenvironment

TNBC

Triple-negative breast cancer

TNF-α

Tumor necrosis factor-α

Treg

Regulatory T cells

TSLP

Thymic stromal lymphopoietin

YAP

Yes-associated protein

Authors’ contributions

Y.L. conceived and organized the manuscript. L.K. and L.W wrote the manuscript, prepared the figures, table, and contributed to the discussion. All authors have read and approved the final manuscript.

Funding

Major International (Regional) Joint Research Program of the National Natural Science Foundation of China, Grant/Award Number: 81920108027, 32470971 and 82271885; National Outstanding Youth Reserve Talent Training Project; Research Capacity Enhancement Project of Chongqing University Cancer Hospital (Y2024003); Funding for Chongqing Young and Middle-Aged Medical Excellence Team.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

All authors agree with the content of the paper and are listed as coauthors.

Competing interests

The authors declare no competing interests.