Dendritic cells have a central role in starting and regulating immune functions in anticancer responses. The crosstalk of dendritic cells with tumors and other immune cell subsets is partly mediated by extracellular vesicles (EVs) secreted by both cell types and is multidirectional. In the case of dendritic cell EVs, the presence of stimulatory molecules and their ability to promote tumor antigen-specific responses, have raised interest in their uses as therapeutics vehicles.
Cancer is a very heterogeneous disease that can develop in almost any tissue due to the tumorigenic transformation of normal cells. Malignant cell transformation is a multi-step, diverse process that might be instigated by genetic factors. Hanahan and Weinberg outlined the six core “hallmarks of cancer” placing the spotlight on the role of the tumor microenvironment in malignant cell progression [1].
Cancer thrives after escaping control checkpoints. The immune system exerts one of the main defense mechanisms against malignant cells via immune surveillance and specific antitumor immune responses. This comprises two arms: innate immune cells that define a first rapid line of defense and adaptive immune cells, which drive an antigen-specific response that includes the development of long-term memory. However, memory traits have also been observed in the innate compartment based on epigenetic and metabolic reprogramming of cells, the so-called “trained immunity” [2]. Among innate immune cells, the main players are phagocytic cells such as neutrophils, monocytes, and macrophages; professional antigen presenting cells (APCs) such as dendritic cells (DCs), or cell slayers such as natural killers (NKs). Adaptive immune cells include antibody (Ab)-producers B lymphocytes; CD8+ cytotoxic T cells; or helper cells, which include the family of CD4+ T cells (Th1, Th2, Th17, Treg, and others). Around and within tumor environment, innate and adaptive immune cells are crucial players [3].
Immature DCs switch to an activated state through a maturation process after stimulation by a danger signal, such as sensing a pathogen-derived molecule, or tissue damage [4]. These cells infiltrate the tumor and recruit effector cells. DCs represent the most effective APCs able to prime naïve T cells and induce an effective antigen-specific antitumor defense. They include a vast variety of cellular types with diverse functions depending on their origin, location, and properties. For instance, DCs can be subdivided into: (i) conventional DCs (cDCs), either resident of lymphoid tissues or migratory, where we can find cDC1 required more for pathogen/tumor immune responses or cDC2 more focus on major histocompatibility complex (MHC)-II based responses; (ii) plasmacytoid DCs, main producers of type 1 interferon (IFN), (iii) tissue-specific DCs such as Langerhans cells (LCs) or dermal DCs (dDCs) and (iv) monocytic-derived DCs (moDCs), producers of tumor necrosis factor (TNF)-α and inducible nitric oxide synthase (iNOS) [5]. In particular, cDC1 are critical in tumor surveillance, antitumor antigen-specific T cell responses, responsiveness to immunotherapies, and are associated with increased patient survival [6].
Extracellular vesicles (EVs) are secreted by cells to the extracellular milieu [7] and comprise mainly three subgroups depending on their origin: (i) shedding vesicles, generated by evagination of the plasma membrane; (ii) exosomes, generated at the multivesicular bodies and secreted by its fusion with the plasma membrane; and (iii) apoptotic vesicles [8][9][10]. These vesicles are constituted by a lipid bilayer containing an assortment of proteins, lipids, metabolites, and nucleic acids, the latter including microRNA (miRNA), mRNA, long non-coding RNA (lncRNA), DNA and mitochondrial DNA (mitDNA) [10][11][12][13]. Although their biogenesis is still under study, there is evidence of an active sorting process of molecules into these vesicles as its cargo is not a mere reflection of the cell content [10][14][15]. For instance, sorting mechanisms include tetraspanins, lipids, specific proteins, post-translational modifications, or endosomal sorting complexes required for transport (ESCRT)-dependent processes [14][16][17][18][19]. Their functions include a variety of cellular processes, but these vesicles are specialized in intercellular communication [20][21]. EVs can function as autocrine, paracrine, or endocrine entering into circulation. Once they reach their final destination, the mechanisms underlying their internalization and signaling processes remain still under consideration [7][8][10]. Almost any type of cell can secrete EVs, including malignant cells and immune cells [8][10][11][21]. There are specific markers for each type of vesicle, specific for a certain cell type or even to distinguish between types of vesicles from the same cell type [21][22][23][24]. Besides, EVs can be detected in any type of biological fluid [21][22]. For example, as EVs from malignant cells convey tumor molecules, they represent good tumor biomarkers and excellent liquid tumor biopsies [25][26]. Interestingly, in cancer patients, the amount of serum-EVs were shown to correlate with a poor prognosis [27]. Vesicle secretion and size heterogeneity has made it difficult to decipher their precise origin and functions, generating some controversy [28][29][30][31][32].
Immune cells can secrete immunologically active EVs [8]. One of the first studies of vesicles in immune responses described the role of B cell-EVs carrying MHC-II molecules on their surface in driving T cell proliferation [33]. Since then, the number of publications on immune cell-derived vesicles functions has not ceased to grow due to their potential in human immunotherapies, for instance against cancer. Among innate immune cells, we will focus our attention on EVs derived from DCs (DEVs) [34]. At the end of the 90s, Zitvogel and colleagues showed that DEVs convey tumor-associated antigens (TAAs) promoting antitumor immunity by effector T cells [35]. In particular, DEVs contain a specific repertoire of molecules like T cell co-stimulatory molecules (CD86, CD80, CD40) [36][37], antigen presenting molecules (MHC-II, MHC-I) [36][37][38], adhesion molecules (Integrins, intercellular adhesion molecule 1 (ICAM-1), dendritic cell-specific intercellular adhesion molecule-3-Grabbing non-integrin (DC-SIGN)) [39], NK modulation molecules (TNF-α, interleukin 15 receptor α (IL-15Rα), NKG2D-L) [40][41] and the EV markers such as tetraspanins (CD9, CD81, CD63), ESCRT complex proteins (Tumor susceptibility gene 101 (TSG101), ALG-2-interacting protein X (ALIX)), heat shock proteins (HSC73, HSP84) and others (SYNTENIN-1, ACTIN) [37]. The amount of MHC molecules or co-stimulatory molecules depends on the physiological state of the DC [38][42][43]. In fact, EVs secreted from DCs can transfer functional MHC-I-peptide complexes to other DCs [44]. Not only DEVs induce stimulation of naïve CD4+ T cells, but also these EVs are also used by mature DCs as a source of tumor antigens [43][45]. Apart from proteins, nucleic acids sorted within DEVs play an important role in the regulation of immune responses, in particular miRNAs both from immature and mature DCs [46][47][48]. The secretion of these vesicles by DCs can be altered upon exposure to different stimuli [49][50]. Highlighting the heterogeneous nature of these vesicles, different subtypes of DEVs could also be found, each of which could perform a variety of functions [50][51].
One of the main functions of DEVs is T cell activation:
- EVs from an antigen-loaded DC bearing tumor antigens may promote CD4+ and CD8+ T cell responses by direct antigen presentation, which leads to tumor growth suppression [35][36][52] with increased efficiency in the case of mature DC-derived EVs [39].
- Moreover, by coating DCs with the pMHC-loaded EVs, a process known as MHC cross-dressing, T cell response is reinforced and amplified [53][54].
- DEVs can also be internalized by other DCs as a source of exogenous peptide-loaded MHC (pMHC), which may be subsequently presented to naïve, primed, or memory T cells [43][44][45][55][56][57]. This process has a special relevance in organ transplantation as acceptor DCs incorporate donor-DEVs to stimulate allospecific T cells [58]. A mechanism and source of tumor antigens is the internalization of EVs containing the full antigen or peptide, that would be presented later by endogenous MHC-I molecules at the acceptor cell in a process called cross-presentation [59]. In fact, EVs loaded with the whole antigen are more efficient in antigen presentation than EVs bearing only pMHC [60][61]. These antigen-loaded EVs elicit a Th1 CD4+ and CD8+ T cell response dependent on B cell activation [62][63]. This response might also be enhanced by the presence of CD80 and ICAM-1 in those EVs [39]. Besides, bystander T cells can promote DC maturation in the absence of innate stimuli [64][65][66], which is also reflected in DEVs supporting subsequent specific antigen T cell activation [67].
- In addition, DCs can incorporate pMHC-loaded vesicles from other cell origins, such as epithelial cells, to potentiate antigen presentation to T cells [68].
Antigen-loaded DEVs can play many other functions in addition to modulating T cell responses, such as promoting humoral immunity [69][70]. DEVs also contain NKG2D-L and IL-15Rα, contributing to NK cell activation and proliferation [41]. DEVs can directly activate NK cells via TNF-α as they display it on their surface together with FasL and TRAIL. These NK cell responses, along with the apoptotic signaling, contribute to tumor cell removal [71]. Furthermore, EVs from heat shocked-activated DCs bear BAT3, the ligand for NKp30, mediating cytokine release and NK cell cytotoxicity [72].
Therefore, DEVs entail an important tool to elicit the immune response against malignant cells. These vesicles can boost both T cell and NK responses, whose cytotoxicity leads to tumor cell killing [34][40]. Other DCs can internalize or be coated with DEVs enhancing the antitumor response [43][54].
Immune evasion is one of the emerging hallmarks for cancer [1]. Tumor cells manage to escape immune responses, particularly inhibiting T cell activation and DC differentiation [73]. Notably, tumor cells can shed a large amount of EVs, highlighting the importance of vesicle secretion during tumor development [74][75]. Tumor-derived EVs (TDEVs) can mediate many aspects of the immune response [76]. TDEVs can alter the microenvironment of the tumor promoting both pro/anti-inflammatory responses on monocytes, macrophages, and DCs, as well as anti-inflammatory responses acting on NKs and Treg, thus modulating angiogenesis, invasion, apoptosis, and metastasis [76][77]. TDEVs contain a repertoire of proteins, DNA and miRNAs that alter DC function and differentiation, and lead to a change in the immune response, thus favoring or hampering tumor progression. Therefore, the design of DC-based immunotherapies must take into account the conundrum that DC per se might fight the tumor, but TDEVs at the tumor microenvironment might sway DCs function into a pro-tumor phenotype.
Tumors have managed to contain a repertoire of mechanisms to evade any antitumoral response. In particular, the importance of TDEVs on these processes is remarkable. TDEVs can hinder many aspects of antitumor immune response, from increasing suppressor cell populations, like myeloid-derived suppressor cells (MDSCs), to decreasing antigen presentation processes. On the contrary, DCs can take advantage of TDEVs as they convey TAAs, which might serve as a source for direct or indirect presentation mechanisms. However, this remains to be fully characterized, especially in the case of direct TAAs presentation carried by DEVs. Finally, diversity in EVs methodology and DC sources used in these studies may pose a confounding factor when interpreting these data.
The fine modulation of the immune responses that EVs are able to perform, as well as their abilities for shuttling different biomolecules, including proteins that may serve as antigens to mount an immune response, have made EVs interesting candidates for different uses in therapeutic and prevention contexts where the immune system has a major role.
A pioneering study using autologous DEVs showed rejection of tumors in mice when loaded with tumor peptides and therefore used as a cell-free vaccine [35]. Driven by these observations, several in vivo and clinical trials have followed and explored the use of EVs (more prominently DEVs) as potential immunotherapeutic agents in cancer (reviewed in [78][79]). In this section, we will focus on the use of DEVs as therapeutic agents in the context of cancer. Importantly, we differentiate between vaccination approaches, when EVs are loaded with tumor antigens in order to elicit (tumor) antigen-specific immune responses, or immunotherapies if they are not loaded with tumor antigens or their therapeutic effects are based on immunomodulation independently on whether they improve a later antitumoral antigen-specific immune response.
Several sources of EVs have been used as potential tumor vaccine candidates. However, DEVs are the most interesting EVs to activate specific immune responses because: (1) they are antigen-presenting platforms as they contain MHC-I and-II or CD1 loaded with tumor peptides as well as co-stimulatory molecules (CD80, CD86) ([37][39][80][81] reviewed in [82]); and (2) because the activation state of their cellular source can be manipulated. Due to the feasibility of large production of DCs for DEVs generation, most studies that explored the use of DEVs in antitumor settings in ex vivo, in vitro and in vivo have used murine bone-marrow-derived DCs (BMDCs) as a source for DEVs production. These DEVs have been loaded with different TAAs to mount specific T cell responses against tumors (examples can be found in Figure 1A). In these cases, antigen-loaded DEVs were able to increase overall survival and reduce tumor growth. These effects were accompanied by a potent activation of both CD4+ and CD8+ T cells responses. Also, evidence has pointed out that the use of activated DC as producers of DEVs increases the eficacy of DEVs when used as cancer vaccines [83]. Owing to this, several studies have explored the use of DEVs coming from DCs stimulated with TLR ligands [84][85][86].
Figure 1. Therapeutic and prophylactic applications of DEVs in cancer. Mice and human DEVs have been used in antitumor studies as therapeutic or prophylactic agents. (A) DEV-based cancer vaccines have been designed by loading DEVs with peptide-loaded MHC (pMHC) molecules, tumor proteins and lysates, or mRNAs that encode neoantigens with the goal to mount antigen (Ag)-specific antitumor responses. (B) DEVs-based anti-pathogen vaccine platforms for cancer prevention have been designed either by natural-occurring antigen loading to DEVs or by fusing antigens to DEV proteins. Also, their ability as adjuvant carriers and antigen-DEV formulations has been explored to increase vaccine efficacy. In addition, DEVs are a potential platform for mRNA vaccine delivery. (C) DEVs have been used as immunotherapeutic agents by exploiting their immune stimulatory properties. DEVs can directly stimulate innate cells such as natural killer T cells (NKT), NK, and γδ T cells or modulate the immune response indirectly by the delivery of miRNAs; or induce tumor cell apoptosis via ligand interaction or targeted delivery of chemotherapeutics. MAGE, melanoma antigen gene; OVA, ovalbumin; E7, human papillomavirus E7 protein; HBV, hepatitis B virus; HBsAg, surface antigen of the HBV; Nef, negative regulatory factor from human immunodeficiency virus; α-GalCer, α-galactosylceramide; poly(I:C), polyinosinic:polycytidylic acid; FasL, Fas ligand; TNF, tumor necrosis factor; LAMP2b–iRGD, lysosome-associated membrane protein 2 (LAMP2b) fused to the integrin-specific peptide iRGD.
In all these cases, DEVs are used as therapeutic rather than prophylactic vaccines. Indeed, the prophylactic use of DEVs against cancer is largely unexplored because intervention is usually used when the disease appears. However, many pathogens have been described to induce carcinogenesis by direct or indirect mechanisms. Hence, implementation of prophylactic vaccines that target oncopathogens constitute a potential health benefit. EVs in general have been explored as a novel vaccine platform for infectious diseases to a lower extent than cancer vaccines, and no clinical trial have been performed to date. In some cases, EVs have shown increased efficacy than traditional vaccine formulations against pathogens. Targeting of pathogen proteins to EVs endowed efficient antigen-specific cellular [87] and humoral immune responses [88]. Other strategies that use DEVs in vaccination are depicted in Figure 1B.
DEVs can modulate T cell activation as DEVs coming from human moDCs that were not exposed to a specific antigen could stimulat CD4+ T cell ex vivo generating a Th1-type response when using small DEVs; and towards Th2 when using large DEVs from immature cells [50]. Also, DEVs can stimulate NK cell functions. Using murine BMDC DEVs administered intradermally to mice, NK cell proliferation and activation that mediated by IL-15Ra and NKG2D ligands was observed [89]. Other immunotherapeutic strategies for the use of DEVs are depcited in Figure 1C.
To date, four phase I and one phase II clinical trials have been performed using DEVs as immunotherapeutic agents.The outcome of this trials in summarized in Table 1.
Targeted Tumor Type |
Phase of Trial |
n 1 |
Treatment |
Loaded Antigen |
Immune Effects |
Clinical Outcome |
Ref. |
NSCLC (stage IIIb and IV) |
I |
13 (9) |
DEVs from moDC |
MAGE-A3, -A4, -A10, and MAGE-3DPO4 peptides + CMV and tetanus toxoid peptide (direct or indirect loading) |
DTH reactivity against MAGE peptides in 3/9. MAGE-specific T cell responses in 1/3. NK lytic activity in 2/4. CMV responses. Increase of Tregs in 2/3. |
Well tolerated. Mild adverse events. Stabilization after progression in 2/9. |
[90] |
Melanoma (stage IIIb and IV) |
I |
15 |
DEVs from moDC |
MAGE3168–176 and MAGE3247–258 |
Specific T cell responses in peripheral blood not detected. One case of tumor infiltration of activated T cells. NK cell number and NKG2D function recovered in 7/14. NKG2D expression in CD8+ T cells in 6/14. |
No toxicity (mild adverse events). One patient exhibited a partial response. |
|
Colorectal cancer (stage III or IV) |
I |
40 |
AEVs + GM-CSF |
Contain CEA |
DTH response as well as a CEA-specific CTL cell response |
Well tolerated. Stabilization in 1 and minor response in 1. |
[92] |
NSCLC (stage IIIb and IV) |
II |
26 (22) |
DEVs from IFNγ-matured moDC |
MAGE-A1, MAGE-A3, NY-ESO-1, Melan-A/MART1, MAGE-A3-DP04 and EBV peptides. |
Tumor antigen-specific T cell responses only in 2/8. Increased NKp30-dependent NK cell functions. |
Stabilization with continuation of injections in 7. Long-term stabilization in 1. Hepatotoxicity in 1. |
[93] |
Table 1. Clinical trials using DEVs performed to date and their main immune and clinical outcomes.
The increased efficacy and versatility of EV-based vaccines have made them potential candidates for rapid development of vaccines against emerging infections [94][95]. For example, EVs targeting has not only increased immunogenicity of EV-based vaccines as discussed before, but it has also been shown to improve the humoral responses in adenoviral vector vaccines, including ChAdOx1, one of the candidates leading the race for a SARS-CoV-2 vaccine [96]. Interestingly, five registered human clinical trials are exploring the use of EVs as therapeutics against COVID-19. Besides, EV-based vaccines may contribute to the new era of mRNA vaccination for emerging pathogens and personalized cancer vaccines via delivery of mRNAs encoding neoantigens [97][98]. Also, the classical rationale behind the use of DEVs as tumor vaccines or vaccines for oncogenic pathogens relies on the inducing antigen-specific immune responses and adaptive memory. However, the broadening of the concept of vaccination with the appearance of the first generation of trained immunity-based vaccines [99] opens new horizons in exploiting this new arm of the immune system as a new source of therapeutic strategies for immunotherapy and cancer vaccines.
In the context of tumor development, EVs are present in the tumor microenvironment and are secreted by a variety of cell types. On the one side, immune cells produce EVs to fight against tumor progression and metastasis. As illustrated before, DCs, in particular, can secrete EVs to increase the T cell response by enhancing antigen presentation by a diversity of mechanisms. On the other side, malignant cells produce EVs to escape immune responses and virtually, they can interact with every type of immune cell. Additionally, the great ability of DEVs as immune modulators and kick-starters of robust antigen-specific T cell responses and NK cell responses have allowed the use of DEVs in different immunotherapeutic settings: both as novel and effective cancer vaccines and cancer immunotherapies (Figure 1). Several in vivo studies as well as clinical trials support the increased efficacy of the use of DEVs as cancer vaccines compared to DC-based vaccines. However, the prophylactic use of DEVs as a novel and versatile vaccine platform against infectious agents, including oncopathogens, is still at its inception. Combining the use of DEVs and emerging technologies such as mRNA vaccination or the exploitation of trained immunity mechanisms will push forward the frontier of tumor vaccination approaches.
This entry is adapted from the peer-reviewed paper 10.3390/cancers12123558