Exosomes, firstly discovered in the early 1980s, represent a class of extracellular vehicles (EVs) at a size of 30~150 nm in diameter that are secreted by all types of cells (1, 2). Exosomes were initially considered as a means for maturing reticulocytes to get rid of superfluous proteins and discard garbage (1, 2). Accumulating evidence has suggested their other roles besides eliminating unwanted molecules from parental cells such as intercellular communications, cell content exchange, immune system modulation, antigen presentation, and pathogen propagation (3–5). With our incremental understanding on exosome, its prominent functionalities under both normal and pathophysiological conditions such as lactation (6), immune homeostasis (7), neuronal signaling (8), and disease development (9, 10) have been acknowledged and taken advantages of for the purpose of theranostics.

A rising momentum has been witnessed on exosome recently, in both academia and industry. Exosome has been integrated into the field of precision medicine as it provides an extremely useful source of biomarkers for cancer diagnosis and an excellent tool for immune-therapeutics and drug delivery (7, 11). The diverse range of biofluids capable of producing exosomes makes exosome-based diagnostics minimally invasive, easy to use, and fast in detection (12, 13). By isolating a patient’s exosomes, modifying them with appropriate nucleic acids or proteins, and transfusing them back to the patient (14, 15), exosomes can be administrated in clinical practice in a similar fashion to adoptive cell therapy (ACT) that is characteristic of extreme personalization. At the same time, exosomes offer a cell-free solution. That is, they can be manufactured ex vivo free of cells and thus be exempted from the risks and difficulties of administering cells to patients (16).

For these beneficial traits, the era of exosomes has arrived. In academia, there had been approximately 20,000 publications on exosomes, among which three quarters have been published within the past 5 years. Clinically, at least 204 clinical trials associated with exosomes have been launched, with 114 and 74 trials related to therapeutics and diagnosis, respectively, and two diagnostic tests (i.e., Bio-Techne’s ExoDx Prostate IntelliScore Test for prostate cancers [17), Guardant360 CDx test for non-small cell lung cancers (18)] have been approved by the FDA. Commercially, at least seven partnership deals, eight large venture capital events, and two landmark acquisitions have occurred in the exosome industry during the past 5 years. Given the surging amounts of scientific publications, the rising number of clinical trials, and the swelling appetite among investors for exosome biotechnology, it is imperative to characterize the features of exosomes (derived from various sources) that enable their diversified innovative theranostic applications and, ultimately, our improved power over diseases including cancers.

Among the varied types of disorders and clinical applications where exosomes may intervene (19, 20), this review focuses on cancer therapeutics. We briefly introduce some basic information on exosomes in Section 2, categorize the primary characteristics of exosomes enabling their onco-therapeutic potential (i.e., immune modulation and cargo delivery) in Section 3, cast our insights on current and emerging innovations relevant to exosome-associated onco-therapeutic strategies in Sections 4 and 5, and focus on the status and challenges in the clinical translation of exosomes as an onco-therapeutic tool in Sections 6 and 7. Importantly, we forecast possible synergies cold atmospheric plasma (CAP), an emerging onco-therapeutic, can create with exosomes toward enhanced cancer control.

The initial form of exosome in eukaryotic cells is early endosome which is tube-like and distributed in the peripheral part of the cytoplasm (3). When early endosome develops into late endosome, it becomes spherical, accumulates around the nucleus, and forms intraluminal vesicles inside the lumen (3). Late endosome either fuses with the lysosome toward content degradation or fuses with the cell membrane to release its intraluminal vesicles in the form of exosomes (3) ( Supplementary Figure 1 ). Exosomes can be secreted either constitutively or in an inducible manner (21, 22). Tumor suppressors such as p53 can activate exosome production and secretion in response to external stimuli such as oxidative or toxic stress (22). It has been demonstrated that irradiation of human prostate cancer cells can trigger DNA damage and thus induce a p53-depenent increase in exosome secretion (23). Once exosomes are taken up by other cells, they can either release their inner content without affecting the membrane integrity or be entirely endocytosed and re-legated to clathrin-coated pits (3).

Exosomes contain many diversified cell surface molecules that allow them to participate in intercellular material exchange. The content of exosomes is complex which contains thousands of proteins and nucleic acids, as well as hundreds of lipids (24). Exosomes are enriched with proteins involved in antigen presentation such as CD1 and major histocompatibility molecules I and II (MHCI/II), co-stimulatory molecules such as CD86, adhesion molecules such as CD11b and CD54, heat shock proteins such as HSP70 and HSP90, cytoplasmic proteins such as Annexins and Rab proteins, membrane proteins such as CD55 and CD59, signal transduction proteins such as G-proteins, and protein kinases (3). Exosomes do not contain mitochondrial, nuclear, and endoplasmic proteins (25). Exosomes derived from immune cells display proteins that pertain their roles in immune responses in addition to typical exosome contents; e.g., exosomes secreted by dendritic cells (DCs) contain CD80 and CD86 for naïve CD4+ T-cell activation (26). Exosomes also contain numerous nucleic acids such as let-7, miRNA-1, miRNA-15, and miRNA-16 which play critical roles in angiogenesis, hematopoiesis, exocytosis, and tumorigenesis (25). Recent studies have reported the critical functionalities of exosome-derived long non-coding RNAs (lncRNAs) in cancer initiation and development. For instance, PTENP1 in exosomes derived from normal cells confers tumor-suppressive roles on breast cancer cells both in vitro and in vivo (27). Compared with the source cells, lipids contained in exosomes are enriched with cholesterol, sphingomyelin, glycosphingolipids, hexosylceramide, lactosylceramide, and phosphatidylserine and contain less phosphatidylcholine, phophatidylethanolamine, phosphatidylglycerol, phosphatidylinositol, and cholesteryl ester (28). Exosomes have asymmetric membrane bilayers regarding lipid decomposition (3). While sphigomyelins, sphingolipids, and most phosphatidylcholines are distributed in the outer leaflet of exosomes, all other types of lipids are mainly concentrated in the inner leaflet (28). Exosomal cargos are involved in various signalings in recipient cells and can modulate diverse biological processes such as autophagy (29) and inflammation (30).

Following successive innovations in creating synergies between exosomes and various drugs and exosome-modulating approaches, relatively little has been focused on the potential aid of emerging technologies in exosome innovations. Here we introduce an emerging anticancer tool, CAP, and its possible synergies with exosomes toward improved onco-therapeutics.

CAP, a fourth state of matter that relies on reactive species toward selective control over malignant cells for death (147–151), has become an emerging onco-therapeutic tool with great translational potential (152, 153). It is a cocktail of multiple reactive oxygen and nitrogen species (RONS) such as short-lived species singlet oxygen (O), hydroxyl radical (OH·), superoxide (O²⁻), and nitric oxide (NO·), and long-lived species hydrogen peroxide (H₂O₂), ozone (O₃), anionic (OONO⁻), and protonated (ONOOH) forms of peroxynitrite. These species, by themselves or their interactions, interact with the surface of cancer cells and generate a series of intracellular signalings that selectively arrest cancer cells at various types of death states (such as immunogenic cell death (ICD) (154), apoptosis (150), cell-cycle arrest (147), autophagy (155), ferroptosis (156)) by perturbating their redox homeostasis. Such a selectivity not only attributes to the higher basal level of cells under the malignant state as compared with their healthy peers but also is associated with the membrane features of cancer cells. The latter includes, e.g., a high expression of aquaporins on cell surface that is associated with increased H₂O₂ uptake (157) and a high local concentration of catalysis on the surface that determines the specific response of tumor cells to self-destruction as a result of secondary singlet oxygen generation (158).

We identify three possible synergies that CAP may create with exosomes. That is, CAP may synergize with exosome toward enhanced immuno-modulation, function as the cargo of exosomes, and boost exosome production.

Clinical trials using exosome as an onco-therapeutic approach can be dated back to almost 20 years ago when two phase I clinical trials using immature DC-derived exosomes for treating melanoma and lung cancer were launched (14, 15). In the first trial, exosomes originated from DCs were pulsed with melanoma-associated antigen (MAGE) and inoculated to 15 stage III/IV melanoma patients, where approximately 62% patients exhibited an enhanced NK-cell activity with no major toxicity reported 2 weeks after immunization (14). In the second trial, DC-derived exosomes were loaded with HLA-restricted MAGEs, followed by back infusion into patients carrying HLA A2+ non-small cell lung cancers (NSCLCs), where one-third of the patients showed MAGE-specific T-cell responses and, among the four analyzed patients, half presented an increased NK-cell activity after a 1-month weekly treatment (15). It has been proposed that mature DC-derived exosomes can induce more potent T-cell priming than those derived from immature DCs (173). A non-randomized phase I/II clinical trial achieved antigen-specific T-cell responses among seven esophageal cancer patients using mature DC-derived exosomes pulsed with SART1 (a biomarker of squamous cell esophagus carcinoma) (174). However, the use of mature DC for exosome generation is not the sole golden standard for activating an antigen-specific T-cell response that also relies on many other factors. For instance, a phase II clinical trial documented the use of exosomes derived from mature DCs in treating patients carrying NSCLCs in 2016. In this trial, mature DC exosomes loaded with MHC class I and II-restricted cancer antigens and IFN-γ were administrated to 22 patients. The results showed elevated NK-cell activities without obvious toxicity; yet, unlike what was expected, no cancer-specific T-cell immune response was boosted (175). This may be attributable to the suppressive role of PD-1 on T-cell activity, the expression of which from exosomes was elevated by IFN-γ. Other studies documented the indispensable role of CD4+ T and B cells in activating cytotoxic T cells by DC-derived exosomes (176).

Besides DC-derived exosomes, clinical efforts in cancer treatment have also been devoted to utilize exosomes originating from other cell sources. One phase I clinical trial used ascites-derived exosomes together with granulocyte-macrophage colony-stimulating factor as a combined therapeutic for advanced colorectal cancers. After a 4-week treatment administration, colorectal cancer patients demonstrated a strong cytotoxic T-cell response against cancer cells carrying the carcinoembryonic antigen (a biomarker of colorectal cancers) (177).

Ongoing efforts have been made to translate exosomes into clinical use for drug delivery. For example, Codiak BioSciences, being one of the major biotech start-up companies dedicated for exosome therapeutic development, constructed an engineered exosome for pancreatic cancer treatment by delivering siRNAs targeting the KRAS (G12D) mutation (iExosomes) (142), for which they have obtained the investigational new drug (IND) approval for conducting the phase IA/B clinical trial from US FDA (NCT03608631).

The many clinically favorable features superimposed on exosomes such as the immune-boosting and molecule-carrying roles have provided us with a unique option to target tumors using cell-free cancer vaccination (7) capable of synergizing with various onco-therapeutic cargoes. This has led to an industrial zest toward its clinical translation that, however, is currently limited by the following therapeutic hurdles ( Figure 3 ).

Large-scale production is essential for the wide clinical application of novel techniques including exosomes. Cultivation and isolation are critical steps in exosome manufacturing. Conventional cell culture has been traditionally used for cultivating source cells of exosomes. The 3D culturing technique (through the use of bioreactors) has been established to maximize the cell culture surface area toward enhanced exosome yield (178, 179) which, however, is not necessarily cost-effective as more media and more frequent passages are needed for bioreactor-based cell culturing. Ultracentrifugation represents a dominant approach for exosome isolation that suffers from quantity loss due to exosome heterogeneity and from biological contamination (180).

How to minimize the heterogeneity of exosomes represents another important problem challenging large-scale exosome production. Among all possible influential factors, the heterogeneity of parental cells (even of the same cell type) imposes the leading effort that may substantially affect the quality, consistency, and functionality of exosomes produced. For instance, exosomes released from the muscle cells of aged mice are prone to induce inflammation and accelerate aging due to the enrichment of miRNA-34a than those derived from young mice (20, 181). Thus, a careful assessment on potential therapeutic indexes of source materials with critical impacts on the functionalities of derived exosomes such as age should be considered as an important guideline for large-scale exosome manufacturing.

Besides issues relevant to exosome mass production, there also exist unresolved problems on how to translate preclinical experiences into clinics. First and foremost is how to determine exosome dosage under different therapeutic scenarios, where experiences gained using cells or animal models need to be extrapolated into the clinical level. Other clinical features such as exosome distribution, uptake, and half-life should be carefully investigated to avoid off-target tissue accumulation prior to clinical administration (182).

Also worth mentioning is the limitations of exosomes in stimulating the immune system. Similar with canonical agents, exogenous exosomes cannot produce long-term consecutive stimulation to the immune system and may lose the immune-stimulatory efficacy if running exhausted. However, it is not feasible to inject overdosing exosomes to the human body as an overactivated immune system may lead to undesirable clinical outcomes such as cytokine storm that, sometimes, can be life-threatening. Thus, achieving extended exosome release by synergizing with materials such as hydrogel may represent a promising solution and a possible future research direction. Another remaining issue is how to determine the dose and frequency of repeated exosome administration as well as the dosage regime that may differ among patients, without which it may be difficult to achieve an optimal or expected therapeutic response. Thereby, reinforcing our control over the choice of exosome dosage under each treatment scenario represents another major obstacle awaiting to be resolved.

Lastly, lack of industrial standards limits exosome commercialization. Exosomes derived from different sources have different features and thus require different manufacturing protocols and quality control standards. Besides, the stability, safety, potency, and quality of exosomes should all be carefully controlled throughout the manufacturing process and under the shipping conditions. Thus, a set of standards systematically covering all these scenarios is urgently needed for standardized exosome manufacturing that requires collective advances in technologies and government/industrial regulations.

Being homogeneous EVs secreted by various types of cells for intercellular communication, exosomes carry information from host cells that can be utilized for diagnosis and, importantly, as immuno-therapeutics due to inherited features from their immune cell origin (for exosomes derived from some immune cells). On the other hand, exosomes are ideal natural nano-carriers for drug delivery due to their small sizes that allow them to penetrate through BBB, flexibilities in surface modulation that allow tissue-specific targeting, endogenous origins that enable their biocompatibilities, natural expression of surface receptors that allow their easy communication with target cells, and bilayer membrane structures that protect their imbedded cargos from, e.g., gastric acidity.

CAP, being an emerging onco-therapeutic approach and redox modulator, can aid in exosome-based onco-therapeutics by sensitizing exosome-based immunotherapies, being the cargo of exosomes for effective malignant cell removal with little side effect, and functioning as a controllable inducer for massive exosome production.

We forecast the emerging and wide application of exosomes in onco-therapeutics, and the prominent role of CAP in availing this process toward, hopefully, eventual malignant tumor eradication. Our insights not only categorize the unique traits of exosomes favorable for disease management (especially for treating brain cancers) but also identify a novel opportunity for cancer therapeutics through a combined use of exosomes and CAP.

XD conceived the study, conducted the literature searching, drafted the manuscript, and prepared the figures. YY contributed in manuscript revision. FH provided the financial support. All authors contributed to the article and approved the submitted version.

This work was supported by the National Natural Science Foundation of China (Grant No. 81972789), Fundamental Research Funds for the Central Universities (Grant No. JUSRP22011), and Technology Development Funding of Wuxi (Grant No. WX18IVJN017). The funding bodies played no role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.

Author XD was employed by CAPsoul Medical Biotechnology Company, Ltd.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.