Due to their rich composition and capacity to interact with other cells, EVs function in many biological processes including the efficient removal and delivery of molecules between cells to modulate intercellular communication, cell maturation and adaption to environmental changes (Buzas, 2023). For example, in the context of the central nervous system, EVs derived from astrocytes appear to facilitate neuroprotection by shuttling neuroglobin to neurons, which exhibits antioxidant, anti-apoptotic and anti-inflammatory effects (Venturini et al., 2019). EVs derived from osteoblasts and osteoclasts also showed the ability to enhance both osteogenesis and osteoclastogenesis to maintain bone homeostasis (Pieters et al., 2019). EVs can control biological functions directly by triggering surface receptors on recipient cells or indirectly by fusing with the plasma membrane and releasing nucleic acids, genes, proteins or infectious agents in recipient cells. Through these actions, EVs have been shown to play an important role in stem cell maintenance (Ratajczak et al., 2006), tissue repair (Gatti et al., 2011), immune surveillance (Raposo et al., 1996), and blood coagulation (Del Conde et al., 2005).

In the context of the immune system, EVs have immunomodulatory effects, as evidenced by the ability to transfer antigens between antigen presenting cells (APCs). Indeed, EVs carry pre-formed functional peptide-MHC complexes from APCs such as dendritic cells to recipient cells, which stimulate maturation and differentiation in immature dendritic cells and monocytes (Wei et al., 2017; Schierer et al., 2018). EVs derived from dendritic cells also promote interferon-gamma (IFN- γ) production by naïve CD4⁺ T cells and trigger the differentiation into T helper 1, T helper 2 and regulatory T (T_reg) cells (Kowal et al., 2016). EVs derived from T cells themselves have also emerged as critical immunomodulators. Small EVs derived from T cells were enriched with mitochondrial DNA (mtDNA), which primed dendritic cells to a specific antigen (Torralba et al., 2018). EVs derived from T cells are also capable of encapsulating immunomodulatory cytokines, enhancing cytokine stability in comparison to their free forms (Fitzgerald et al., 2018). The stabilizing effect of EVs on the encapsulated cargo has been exploited for drug delivery and biomarker applications (see sections 1.3.2 and 1.3.3).

EVs have also been implicated in triggering inflammatory pathways. Blood-derived EVs from patients with dermatomyositis contained double stranded DNA that stimulated the STING pathway to induce pro-inflammatory cytokine secretion (e.g., type I interferon) (Li et al., 2021). EVs have also been shown to influence the polarization of macrophages to enhance inflammation. EVs enriched with miRNA-155 are abnormally elevated in inflammatory bowel disease, acute lung injury, myocardial ischemia reperfusion injury and triggers macrophage polarization toward an inflammatory (M1) phenotype (Tang et al., 2022). The development of inhibitors that target EV-mediated inflammation and disease progression could be a new approach for treating inflammatory diseases.

A multifaceted role for EVs is reinforced with their ability to regenerate/repair tissue and damaged organs. EVs can transfer microRNAs, messenger RNAs and proteins, which stimulate mitogen-activated protein kinase, Wnt/β-catenin, PI3K/Akt, Notch, TGFβ/Smad, STAT and Hedgehog signalling, which have been linked to enhanced proliferation, migration and survival of recipient cells (Nagelkerke et al., 2021). Notably, EVs have also been shown to downregulate inflammation, attenuate cytokine storm, and promote angiogenesis, as well as trigger lineage-specific cell differentiation, which are critical to promote functional tissue repair/regeneration (Nagelkerke et al., 2021).

Since EVs can be secreted by all cells in the body in response to specific conditions (both physiological and pathological), EVs have been proposed as promising novel biomarkers for numerous diseases. EVs possess competitive advantages compared to traditional biomarkers, including (i) the ability to reflect disease progression and monitor treatment response via vesicle origin and their loaded cargo, (ii) their structure, which protects encapsulated cargo during long-term storage and freeze/thaw cycles; and (iii) the stability of EVs in body fluids (Lai et al., 2013; Tian et al., 2020). Therefore, EVs may have utility to predict flares in diseases that fluctuate over time such as, autoimmune diseases Sjögren’s syndrome, systemic lupus erythematosus and systemic sclerosis (Cecchettini et al., 2019).

In the context of diagnosis of autoimmune diseases, numerous tests can be conducted on biological fluids to examine autoimmune markers, such as autoantibodies and inflammatory signals. However, identical markers can be presented across numerous autoimmune disorders, which can make diagnosis uncertain and lack specificity (Castro and Gourley, 2010). Circulating EVs may uncover tissue-specific autoimmune-associated protein expression to improve diagnostic specificity (Cecchettini et al., 2019). An increase of platelet-derived EVs and endothelial microparticles was found in patients with Sjögren’s syndrome compared to healthy controls (Bartoloni et al., 2015). Another study found that 36 proteins associated with activation of the innate immune system (i.e., SIRPA and LSP1), were upregulated in EVs from the saliva of patients with Sjögren’s syndrome in comparison with healthy controls (Aqrawi et al., 2017). Other studies showed that hsa-miR-768-3p and hsa-miR-574-3p in the EVs derived from saliva can be utilized as biomarkers for inflammation and salivary gland dysfunction in Sjögren’s syndrome (Michael et al., 2010; Alevizos et al., 2011).

Due to their unique properties obtained from their originating cells, EVs can also be used to monitor disease progression and therapy in real time. In cancer, EVs represent relevant tumour constitutes during each stage of cancer management. For instance, EVs have been used recently as a predictive biomarker for outcome of checkpoint inhibitor therapy, anti-PD-1 (programmed death-1) (De Rubis et al., 2019). Additionally, cancer-derived EVs carrying functional programmed death-ligand 1 (PD-L1) on their surface could serve as biomarkers for therapy resistance (Chen et al., 2018).

EVs possess an inherent capability to target specific regions, tissues, and organs (Walker et al., 2019). Harnessing this capability would enable targeted delivery of therapeutics and thus reduce the side effects due to off target delivery and toxicity associated with conventional vehicles such as synthetic nanoparticles (liposomes, polymeric nanoparticles) (de Jong et al., 2020). It has been shown that the accumulation of liposomes at high doses in macrophages can influence their phagocytic activity, resulting in immunosuppression and increasing the risk of infections (Sercombe et al., 2015). In stark contrast to conventional delivery vehicles, EVs exhibit low toxicity and low immunogenicity, can penetrate blood vessels, and enter the extracellular matrix, and have intrinsic ability to cross physical barriers (e.g., blood-brain barrier) (Murphy et al., 2019a).

EVs contain unique composition that can interact specifically with selected target cells and allow the transfer of cargo to recipient cells. The protein contents and membrane protein influence the targeting behaviour of EVs. For instance, EVs expressing integrin α₆ in complex with subunits β₁ and β₄ are preferentially targeted to S100-A4-positive fibroblasts and surfactant protein C-positive epithelial cells in the lungs (Hoshino et al., 2015). Meanwhile, EVs exhibiting subunit β₅ are directed to Kupffer cells in the liver and EVs expressing β₄ are targeted to CD31-positive endothelial cells in the brain (Hoshino et al., 2015). The lipid composition of EVs also alters their targeting behaviour, the strongest evidence being the targeting of macrophages via detection of phosphatidylserine on the surface of EVs. Phosphatidylserine-containing liposomes have been found to decrease EV uptake by competing with and thereby obstructing EV uptake by macrophages (Matsumoto et al., 2017).

Another important characteristic of EVs is their intrinsic capability to cross biological barriers such as the blood brain barrier, which is particularly difficult to breach with synthetic drug delivery vehicles. Alvarez-Erviti et al. demonstrated that intravenous injection of exosomes isolated from dendritic cells loaded with exogenous siRNA can knockdown BACE1, a gene implicated in Alzheimer’s disease, in particular areas of the brain (Alvarez-Erviti et al., 2011). Moreover, repeated administration of EVs did not decrease the efficacy of gene knockdown, as often seen for viral particles due to their detection by the immune system (van Haasteren et al., 2018). Further evidence showed that brain-derived EVs can be detected in peripheral blood, demonstrating that EVs can cross the blood brain barrier (Vandendriessche et al., 2020). Nevertheless, surface modifications for exosomes should be implemented to improve the targeting ability and achieve brain delivery after systemic administration (Tian et al., 2018). In addition, EVs derived from mammalian cells appear to be biocompatible across species. A previous study demonstrated that intravenous administration of bovine milk derived EVs to mice resulted in moderate cytokine release and no other adverse effects (Somiya et al., 2018). EVs are non-replicative and non-mutagenic, exhibiting low immunogenicity and low toxicity. Moreover, their biological barrier permeability and intrinsic cell targeting make EVs promising candidates for drug delivery systems.

Nevertheless, few aspects of the fundamental biology of EVs remain to be elucidated. There has been relatively little work considering the presence of “corona” on the surface of EVs that can alter interactions with recipient cells (Monopoli et al., 2012). Protein corona form readily on the EV surface (Palviainen et al., 2020), and the enzymatic removal of the corona from EVs has been shown to inhibit the cellular uptake of EVs (Escrevente et al., 2011). While corona clearly impacts on the fate of EV in biological systems and may have a pivotal implication on mechanisms of action of EVs, this area has only recently acknowledged in EV research and the studies on EV-corona remain in its infancy.

Additionally, the routes of EV uptake are highly divergent and depend on EV subtype and cell source (Mulcahy et al., 2014). It is possible that a specific uptake route may result in a greater amount of delivered cargo to the recipient cells than other routes. Elucidating this uptake pathway may increase the efficacy and efficiency of EV-based therapeutic strategies. Other challenges related to the translation of EVs based therapeutics to clinic including EV preparations, scalability, analytics, and regulation will be discussed below (Section 1.5).

For the cell-based loading of cargo into EVs, numerous cellular preconditioning strategies have been applied to modify the composition of released EVs. Incubation and transfection of drugs are two commonly used strategies to load the cargo into the parent cells. For example, a small drug doxorubicin (DOX) has been successfully loaded into EVs derived from pancreatic cancer cells (PCCs), pancreatic stellate cells (PSCs) and macrophages by incubating these cells with DOX for 48 h (Kanchanapally et al., 2019). All the cells released the DOX-loaded EVs, but PCCs achieved the highest efficacy, which could be due to the increased efficiency of cancer cells to efflux drugs compared to other cell types (Borst, 2012). Direct transfection is another method used widely for loading miRNA into EVs (Han et al., 2021). One study showed that transfected HEK 293T cells with miR-21 released EVs enriched with miR-21, which decreased the viability of glioma cell lines (U87-MG, C6) (Monfared et al., 2019).

Additionally, it should be noted that preconditioning the parent cells in different environments can alter the cargo of released EVs. For example, ischemic preconditioning of mesenchymal stem cells (MSCs) resulted in the secretion of miR-22-enriched exosomes, which decreased cardiomyocyte apoptosis, fibrosis and enhanced cardiac function in a murine model of myocardial infarction (Feng et al., 2014a). Furthermore, cardiac stem cells preconditioned with heat shock released exosomes enriched in heat shock factor 1, which were delivered to cardiomyocytes and decreased apoptosis in a mouse model (Feng et al., 2014b). Culturing cells on different substrates or matrices may also affect the cargo within EVs. Indeed, a high abundance of the angiogenic factors VEGF-A and IL-8 were contained in EVs derived from human bone marrow-derived MSCs that were cultured in hollow-fibre bioreactors (Gobin et al., 2021). Another recent study showed that MSCs treated with bioglass, which provided calcium and phosphate ions to the cells, produced EVs with different composition compared to EVs derived from untreated MSCs (Wu et al., 2021). Specifically, the level of miR-342-5p was downregulated and miR-1290 was upregulated in EVs derived from bioglass treated MSCs, and these EVs showed an enhanced capability of promoting vascularization and intradermal angiogenesis of endothelial cells.

Exogenous nucleic acids (siRNA, miRNA, mRNA) or small drugs can be directly inserted into EVs using different approaches, such as electroporation, sonication, chemical transfection reagent (lipofection), co-incubation, or calcium chloride (Li et al., 2018). A recent study showed that an anti-KRAS^G12D siRNA was successfully loaded into MSC-derived EVs by electroporation and mice treated with the resulting EVs showed enhanced survival in a model of pancreatic cancer (Mendt et al., 2018). Another study successfully loaded siRNA into EVs by sonication, and these EVs induced a 50% knockdown of a target oncogene (Lamichhane et al., 2016). SiRNA was also loaded into EVs by lipofection, these EVs caused selective silencing of the RAD51 gene in recipient cells, which resulted in reproductive cancer cell death (Shtam et al., 2013). However, a major drawback of this approach is that the function of exogenous nucleic acids within EVs might be compromised when taken up by the recipient cells as electroporation can introduce precipitation and aggregation of siRNA; this might limit the application of this approach (Kooijmans et al., 2013).

Notably, the potential of EVs to change a paradigm in drug delivery is evident in a billion-dollar partnership between Evox Therapeutics Ltd. and Eli Lilly Co. to use exosome technology to deliver RNA-based therapeutics to the central nervous system. In summary, EVs are promising delivery vehicles for precision medicine that could enhance the therapeutic efficacy of different molecules, while reducing potential side effects and thus ultimately improving quality of life for patients.

There is currently no standardised protocol to isolate and purify large quantities of EVs for clinical use. Research laboratories have utilized different approaches to isolate, purify, quantify, and functionally characterise EVs (Sáenz-Cuesta et al., 2015). Since different sub/populations of EVs are generated depending on the isolation method used, it is vital to standardize protocols for EV isolation and characterisation (Phan et al., 2021). Furthermore, it must be noted that the final product of current isolation methods do not produce pure EVs but rather an EV-enriched formulation containing other components, including high-density lipoprotein (HDL) and low-density lipoprotein (LDL) (Webber and Clayton, 2013). Moreover, EVs derived from blood samples are likely to be contaminated with chylomicrons, which overlap the size distribution and density of EVs making them indistinguishable by current isolation and characterisation methods (Karimi et al., 2018).

Most of the traditional methods to isolate EVs involve multiple steps and procedures, which are labour-intensive. The recent incorporation of tangential flow filtration (TFF) and chromatography-based methods have advanced large-scale production of EVs with current Good Manufacturing Practice (cGMP) (Watson et al., 2018). However, a fundamental understanding of how the substrates used for cell attachment and expansion (e.g., microcarriers) influence cell behaviour and EV composition are necessary to establish reliable protocols and consistency for EV production (de Almeida Fuzeta et al., 2020).

For EV-based therapeutics to progress into clinical applications, selection of an appropriate EV source is required. MSC are an attractive source due to their low immunogenicity and ability to regenerate tissue (Dörnen and Dittmar, 2021). Treatment of MSCs with N-methyldopamine and norepinephrine, is a potential strategy to increase the production of MSC-derived EVs and enhance EV yield (Wang et al., 2020). Nevertheless, the impacts of these strategies on EV composition, efficacy and biological functions associated with their therapeutic use are currently unknown.

Due to the complexity in EV isolation and the heterogeneity of EV samples, it is necessary to reduce pre-isolation and pre-analytical variables through consistent sample collection, storage conditions, and handling procedures.

Regarding the isolation of EVs from cell culture, serum-free medium or EV-free medium are recommended to avoid contaminants from growth supplements such as fetal bovine serum (FBS) (Lehrich et al., 2021). For the isolation or detection of EVs in blood, it is necessary to use a needle gauge that reduces shear forces to prevent platelet activation and secretion of platelet and erythrocyte-derived EVs (Coumans et al., 2017). It is also imperative to dispose the first few millilitres of blood collection, where the cell debris can contaminate the EV isolate (Coumans et al., 2017). The status of patients (fed/fasting) and blood collection time (morning/night) can also influence EV concentration and profile (Danielson et al., 2016). However, further research is required to define the optimal parameters before any recommendations are made (Danielson et al., 2016).

Most studies suggested using plasma for EV isolation from blood, as serum consists of platelet-derived EVs which can co-contaminate plasma-derived EVs (Witwer et al., 2013). Moreover, the results can be influenced by the selection of anticoagulant (heparin, citrate, EDTA) (Yokota et al., 1999). Precautions must be taken at every step of EV production, from initial centrifugation to reduce the processing time and prevent sample degradation due to temperature and enzymatic activities (Gandham et al., 2020).

The storage of EV preparations is challenging, since EVs may be unstable in the carrier media and their biological properties could be compromised by long-term storage (Sivanantham and Jin, 2022). Few studies demonstrated that −80°C is best to preserve EV contents however, repeated freeze-thawing can result in EV aggregation and lysis, causing overestimation of EV size, loss of cargo and biological functions (Jeyaram and Jay, 2017; Maroto et al., 2017; Welch et al., 2017; Théry et al., 2018). There is an urgent need to investigate how various reagents, including storage vials, involved in EV preparation and affected the biological functions of EVs and how to optimise EV stability (Théry et al., 2018).

Currently, there are no regulatory guidelines specific for EV-based therapeutics (Phan et al., 2022). However, the manufacturing and safety requirements for cell- and tissue-based therapeutics could serve as examples for the development of EV-based products (Soekmadji et al., 2020). It is anticipated that the development of EV-based therapeutics is highly dependent on the choice and characterisation of the origin of EVs (Soekmadji et al., 2020). Hence, the current considerations of safety and toxicity applied for cell-based therapies could also be employed for EV-based therapeutics (Lener et al., 2015).

EV-based products are regulated broadly as biological medicinal products; the active substance of which is produced or isolated from a biological source (Silva et al., 2021). The subcategorization of EV-based products is likely to be defined by their active components and complexity (Lener et al., 2015; Silva et al., 2021). In 2018 and 2022, the committee for advanced therapy (CAT) suggested that EVs comprising recombinant RNAs should be considered gene therapy products because the effects of the EV directly correlated to these RNAs (Silva et al., 2021). EVs comprising recombinant proteins or peptides would belong to a different subcategory. Therefore, the knowledge of active substance and mode of action is required for regulatory classification of EV-based products and should be a central focus during product development (Gimona et al., 2021). The technology readiness levels (TRLs) scale, developed by NASA and employed by the US Department of Defence to enable clinical translation while mitigating risks of therapeutics, could be utilised to inform guidelines for the development of EV-based products (MARVEL, 2020), https://cordis.europa.eu/project/id/951768.

Amyloidosis is caused by deposition of misfolded and aggregated amyloid proteins in vital organ systems including the heart, kidney and liver (Merlini, 2017). Amyloid proteins are formed from soluble proteins, which undergo misfolding and consequently, assemble into insoluble fibrils (Almeida and Brito, 2020). These fibrils are non-branching and typically 8–12 nm in diameter (Juneja and Pati, 2020). The deposition of amyloid may be localised, with the deposits occurring in a single organ/tissue or systemic where numerous organs and tissues in the body are affected (Juneja and Pati, 2020).

Amyloidosis can present with a broad range of nonspecific and unpredictably progressive symptoms depending on where the amyloid fibrils deposit (Chee et al., 2010; Lu and Richards, 2019). Evaluation of two independent cohorts of patients with light chain amyloidosis reveal a wide range of affected organs in a majority of patients (% of patients indicated), including heart (70%), kidney (60%), liver (20%), gastrointestinal (15%), peripheral (15%) or autonomic nervous system (10%), and soft tissue (10%) (Wechalekar et al., 2013; Muchtar et al., 2019). There is an urgent need to develop an effective tool that can predict the targeted organ in early stage and prevent progressing pathology and organ dysfunction.

Amyloidosis can be formed from either an abnormal elevation in the concentration of proteins with an acquired or hereditary mutation, or precursor proteins with an intrinsic susceptibility to misfolding or proteolytic remodelling into amyloidogenic fragments (Chiti and Dobson, 2017). Amyloid diseases are classified based on the protein that forms amyloids. Currently, amyloids can be formed from 36 proteins in humans (Sipe et al., 2016).

The most popular types of amyloidosis are immunoglobulin light chain amyloidosis, transthyretin amyloidosis and reactive amyloidosis, the latter occurring secondary to a chronic inflammatory disorder such as infections or rheumatoid arthritis (Picken, 2020). With regards to naming conventions, the amyloid is termed A, followed by an abbreviation of the precursor protein, e.g., AL (amyloid derived from immunoglobulin light chain), ATTR (amyloid derived from transthyretin).

There is an unmet need for an early and accurate diagnosis and typing of amyloidosis. Many different types of amyloid diseases have a similar clinical presentation but require different treatments. For example, most AL treatments are based on suppressing the plasma cells responsible for producing toxic light chains with chemotherapy (bortezomib, cyclophosphamide) and immunosuppression (dexamethansone) (Bianchi et al., 2021) or more recently, anti-plasma cell monoclonal antibody (daratumumab) (Kastritis et al., 2021). In contrast, ATTR is treated with agents that stabilise circulating transthyretin tetramers (tafamidis, diflunisal) to prevent the formation of toxic monomers (Nuvolone et al., 2022) and recently, TTR gene silencing (patisiran, inotersen) to reduce total circulating transthyretin (Schmidt et al., 2022). To ensure patients receive the appropriate treatment, a comprehensive approach with biomarkers, multimodal imaging and invasive tissue biopsy is often required to screen, diagnose and differentiate amyloidosis (Jung et al., 2022).

While a monoclonal gammopathy of unknown significance (MGUS) is a potential biomarker for AL, MGUS occurs in 39% of people with ATTR (Phull et al., 2018), 3% of people over 50 years and is a known precursor for other serious indications like multiple myeloma (Kyle et al., 2018). Moreover, in some cases of AL, routine laboratory tests are insufficient to detect the light chains because the light chain burden is very low, and the titres are similar to healthy donors (Goldis et al., 2023). Without a clear biomarker to distinguish those with ATTR and MGUS (where the paraprotein is not contributing to pathology) from AL (where the paraprotein is causing pathology), patients are subjected to more invasive diagnostic procedures. Late and inaccurate diagnosis of amyloidosis resulted in a low rate of survival after diagnosis confirmation with 6 months median survival for late-stage AL amyloidosis (Sabbour et al., 2021).

Given that EVs are capable of transmitting cellular material specific to dysfunctional cells, tissues and organs, EVs are promising biomarkers for the diagnosis, typing and prognostication of amyloid diseases (Figure 4). The evaluation of EVs as diagnostic biomarkers in amyloidosis is a relatively recent research endeavour with the most extensive investigation in Alzheimer’s disease (AD). Amyloid-β (Aβ) and hyperphosphorylated tau play a role in the pathogenesis of AD and amyloid PET imaging is part of the diagnostic process (Rajmohan and Reddy, 2017). Previous studies demonstrated that EVs facilitated the accumulation of AD-associated toxic cargo. For example, neuronal-derived EVs (nEVs) were detected in the blood and carried Aβ, phosphorylated tau and synapse-related proteins, which could be used to diagnose AD (Li et al., 2022). Additionally, nEVs expressed in individuals in the symptomatic stages of AD contained higher levels of amyloid-β with 42 residues long (Aβ₄₂) (Fiandaca et al., 2015; Goetzl et al., 2016; Winston et al., 2018; Jia et al., 2019; Zhao et al., 2020), tau phosphorylated at threonine 181 (p-tau-181) (Fiandaca et al., 2015; Goetzl et al., 2016; Winston et al., 2018; Jia et al., 2019), total tau (t-tau) (Fiandaca et al., 2015; Jia et al., 2019; Nam et al., 2020), and tau phosphorylated at serine 396 (p-S396-tau) (Fiandaca et al., 2015; Goetzl et al., 2016; Winston et al., 2018) compared to healthy controls.

nEVs have also shown potential as early biomarkers of AD. Fiandaca et al. demonstrated increased levels, relative to healthy participants, of p-tau-181, p-S396-tau, and Aβ₄₂ in plasma-isolated nEVs as early as 10 years prior to the diagnosis of AD-related cognitive damage (Fiandaca et al., 2015). Another study performed longitudinal analysis on 887 plasma samples (over 3.5 ± 2.3 years before converting to AD) from 350 participants, including 128 participants developed AD by the final follow-up and 222 matched controls who continued cognitively normal (Kapogiannis et al., 2019). Individuals with future AD had higher levels of p-tau-181 and p-tau-231 in nEVs than matched controls (who remained cognitively normal over a same length of follow-up).

Similar to nEVs associating with AD, urinary derived EVs may have applications in renal amyloidosis. Urinary derived EVs (uEVs) have been shown to contain immunoglobulin light chain (LC) oligomers–a transition stage from monomers to amyloid fibrils (Šarić et al., 2014). Importantly, LC oligomers in uEVs were specifically detected in active patients with AL but not patients with multiple myeloma or MGUS (Ramirez-Alvarado et al., 2017). Ramirez-Alvarado and colleges also showed that LC oligomers could be found in uEVs derived from patient’s urine with unsolved renal progression despite hematologic complete response, which was defined as a normal serum free light chain ratio and undetectable monoclonal protein by serum and urine immunofixation (Ramirez-Alvarado et al., 2017). These data demonstrated the increased sensitivity of uEVs to detect amyloid LCs over conventional methods and the potential of uEVs as an improved measure of treatment response in AL. Based on these promising proof-of-concept studies, multiple strategies are being developed to overcome the technical barriers (sample standardization, inadequate starting material, etc.) to facilitate the utilization of uEVs as a biomarker for AL disease activity (Cooper et al., 2022).

EVs have been proposed to promote the aggregation and deposition of amyloid proteins. A recent paper showed that aggregation of an inherited transthyretin variant V30M (TTR V30M), which causes familial ATTR, is enhanced on EV membranes (Yamaguchi et al., 2022). Moreover, EVs promoted the deposition of TTR V30M aggregation in specific tissues and organs. It was suggested that the unique membrane proteins presented on the surface of EVs were responsible for directing the deposition of TTR amyloids to specific tissues. However, a detailed analysis of TTR aggregation and EV uptake by recipient cells in affected organs are needed to fully elucidate the pathogenic role of EVs in familial ATTR and whether this pathway applies to non-familial (wild-type) ATTR.

Another recent study demonstrated that EVs derived from vascular smooth muscle cells (VSMCs) senescence are key mediators of medin accumulation (Whitehead et al., 2023). Medin is a 50-amino acid peptide that contributes to aortic medial amyloid (AMA)–the most widespread localized human amyloid. The release of EVs was promoted in the senescence of VSMCs via upregulation of sphingomyelin phosphodiesterase 3 (SMPD3). Senescence of VSMCs also enhanced the accumulation of medin in the extracellular matrix (ECM) by promoting medin aggregation on the surface of EVs and the binding of EVs to heparan sulfate proteoglycan 2 (HSPG2) in the ECM. Further study is needed to determine whether the interaction between EVs and HSPG2 impacts the role of HSPG2 in controlling ECM organization and stabilization via its interactions with various ECM components (Guilak et al., 2021). It remains to be determined whether the pathogenic mechanisms uncovered for EVs in familial ATTR and AMA are generalisable to other types of amyloidosis.