Physiological roles of EVs
Extracellular vesicles (EVs) play an important role in intercellular communication in and between different tissues of an organism [1, 2]. EVs include exosomes and microvesicles. Exosomes are membrane vesicles measuring 60-150 nm that are released into the extracellular space by multivesicular bodies (MVB) after fusion with the plasma membrane. In contrast, microvesicles (ectosomes) budd from the plasma membrane. Both vesicular entities are loaded in different ways in the cell and thus probably have different biological functions. EVs carry complex biological information consisting of DNA, RNA, proteins and lipids that can alter the phenotype of the recipient cell at several levels. The recipient cells can be affected via direct receptor binding, the fusion of their membrane with the EVs membrane and then the release of their encapsulated molecules (transcription factors, oncogenes, miRNA (miR) and long non-coding RNA) [1, 2] [Figure1].
Moreover, the lipid membrane of EVs surrounds and protects their content from degradation and thus allows their physiological and pathological information to be send over a long distance. In this way, EVs are involved in every aspect of the physiology of the body.
In the past, EVs were thought to embody the cellular waste removal system, which is why EVs were commonly named the “cellular trash system”, underestimating their physiological or pathological functions. However, EVs have been implicated in cell-cell communication and have been observed to transfer functional nucleic acid and proteins between cells. In normal physiology one major aspect of EVs seems to take place in the interaction with the immune system. Their immunogeneic properties range from their involvement in antigen presentation enabling EVs to induce a T helper cell response as well as their capacity to activate cytotoxic T cells during infection.
EVs in cancer
Despite their multifaceted roles in normal physiology, EVs maintain and influence essential processes in the pathogenesis of various diseases, and their role in tumor biology is under thorough investigation. It seems, that during oncogenesis, cells increase the production of EVs, probably due to their increased metabolism. The release of EVs by tumor cells is thought to play a major role in facilitating signalling to surrounding tumor cells and to distant sites via blood or other biological fluids. Tumor cell EVs (tEVs) are involved in the establishment of all hallmarks of cancer including tumor growth , affect the tumor's immune escape by modulating T cell activation  build pre-metastatic niches  remodel the extracellular matrix  and promote angiogenesis  [Figure2].
Characterization of the genome and RNA transcriptome revealed that tumor cell subpopulations with different genomic and transcriptomic subtypes can coexist and have emerged from a distant related precursor, most likely a plastic stem-like cell. This intratumoral heterogeneity provides the fuel for treatment resistance and allows cancer to be a dynamic disease [5, 10]. Heterogeneous tumor cells can transfer tEVs to other tumor cell subtypes enhancing their pro-tumorigenic behaviour. Furthermore, analyses of the Cancer Genome Atlas (TCGA) revealed that EV-related signatures are associated with decreased patient outcomes which may be propagated throughout the tumor via EV communication .
Novel strategies that aim to prime the patient’s immune system against cancerous tissue have recently gained momentum in the clinic. There are many reports demonstrating that immune cells can be regulated by tEVs enabling an immune evasion by the tumor. Raposo and coworkers demonstrated first the immunoregulatory potential of EVs, as they showed that EVs secreted by B cells can carry the major histocompatibility complex (MHC-II) . Many more studied followed, leading to an active area of investigation in the field of EV-based immunomodulation, especially in the cancer biology. An emerging general concept is that EVs from healthy immune cells can stimulate the immune system and induce an anti-tumor effect, while tEVs most commonly provoke an inhibition of the immune system. For instance, EVs from antigen presenting cells (APC) can active T-cells and maintain their activation. tEVs on the other hand have been shown to suppress the activity of T-cells , natural killer cells  and enhance the activity of myeloid-derived suppressor cells . Regarding therapy approaches, in vivo data using dendritic cell (DC) EVs demonstrated that DC EVs could be used to induce an antigen specific immune response. The findings stimulated the investigation of DC EVs as an autologous cancer vaccine . In this regard, early mouse studies showed good response rates  that lead to phase I trials in humans for patients with non-small lung cancer and melanoma. In some patients DC EVs promoted disease stabilisation, yet the utility of DC EVs for immunotherapy approaches needs to be established in future trials .
While expending in the primary organ, tumor cells regularly shed tEVs in the circulation. Although EVs cannot survive long in the blood stream, they can protect and deliver their cargo to numerous healthy cells and thereof disturb the delicate balance of specific microenvironment. In melanoma, tEVs expressing Met tyrosine kinase can be found in the bone marrow which provoke the bone marrow stem cells relocation to the lung where they differentiate into blood vessels creating a perfect environment for the colonization of melanoma CTCs . Emergence of pre-metastatic niches can also result from the accumulation of tEVs at a specific site where the interaction of integrins transported by tEVs modify the extracellular matrix (ECM) and the release of growth factors when taken up by nearby healthy cells . tEVs from colorectal cancer can also increase liver macrophages infiltration and polarization to induce a pro-inflammatory phenotype through miR-21 and the release of pro-inflammatory cytokines and chemokines to re-model the ECM for CTCs arrival . Recent investigation in a mouse model of breast cancer showed that tEVs secreted by resistant cells post chemotherapy carry Annexin VI which allows them to accumulate into the lungs where they are taken up by monocytes and endothelial cells leading to increase angiogenesis and formation of pre-metastatic niche post therapy . Therefore, the capacity of tEVs to migrate and accumulate at specific sites favours the development of pre-metastatic niches that precede the arrival of CTC and the emergence of metastases.
A hallmark of cancer is the induction of angiogenesis to secure the supply of nutrition for tumor cell growth and its increased metabolism. tEVs have been shown to act as key regulators of tumor vascularization via transfer of pro-angiogenic molecules from tumor cells to endothelial cells . In various cancers, tEVs have been shown to promote endothelial cell migration, growth and tube formation. This effect was even more pronounced by tEVs produced from hypoxic tumor cells , underlining that tumor cells can alter the cargo of tEVs for their specific needs. More recently, Chen et al. demonstrated that tEVs from breast cancer cells can promote aerobic glycolysis by the transmission of a myeloid specific HIF1a-stabilizing long noncoding RNA to macrophages inducing a feedback loop and lactate release in tumor cells .
EVs as biomarkers in cancer
Besides their implication in tumor promotion and maintenance, tEVs are released into the bloodstream and distributed throughout the body. tEVs end up in almost all body fluids and organs, including blood, saliva, urine, cerebrospinal fluid, bile, breast milk and stool . It is now widely accepted that the molecular cargo of tEVs is representative of the secreting cell and that these molecules are protected from fragmentation and degradation making circulating tEVs ideal for defining subgroups, stratifying patients, and monitoring therapy by "liquid biopsy " . For example, one of the first reports by Skog et al. reported that serum EVs from patients with malignant glioma carry the mutant tumor-specific EGFRvIII protein and contain transcripts coding for this mutant tumor-specific variants [22, 23]. In addition, EVs in CSF and serum may contain mutant IDH1 transcripts . This phenomenon has been demonstrated in multiple cancer types as well as divergent molecular entities, namely miRNA, mRNA and lncRNA as well as DNA . As biomarkers, serum and plasma EVs are indeed even better suited than circulating tumor cells (CTCs) because they reflect the heterogeneous tumor composition and thus the tumor as a whole better than individual CTCs . In total, there are multiple studies highlighting the potential of EVs being a rich and readily accessible source of cancer biomarkers .
Yet, it needs to be mentioned, whilst tEVs represent a promising class of circulating biomarker, that current challenges need to be addressed and hopefully solved within the upcoming years. One major challenge for the filed remains the lack of standardisation of protocols for EV isolation, enrichment and characterisation as well as documentation. To address these challenges efforts are being made to facilitate inter-study methodological comparisons and to develop study guidelines as well as reporting in EV research. A couple of examples are: minimal informations for studies on EVs (MISEV) guideline 2019; EV-METRIC and EV-TRACK (http://evtrack.org).
It is now widely accepted that EVs participate in numerous pathological and physiological mechanisms, and as such they have been intensively studied over the past decade. There is a growing body of evidence that tEVs carry tumor-representative cargo and that they facilitate tumor growth maintenance and dissemination. However, since EVs display a heterogeneous biological entity, the complexity and challenges associated with EV research remains, which needs to be addressed since tEVs are promising candidates for liquid biopsy approaches.