In Part I, the quality-control framework for stem cell-derived extracellular vesicle (EV) products was examined from the perspectives of cell-bank management, product identification, and physicochemical characterisation. The discussion considered the requirements outlined in the Guidelines for Clinical Research Filing of New Technologies for Cell Component and Derivative Therapies (1st Edition) (National Health Commission, 2026), together with the quality assessment framework proposed by Na et al. (2026). Particular attention was given to the use of general EV markers, particle-size distribution, and Zeta potential, highlighting both the common principles and differences between the two approaches.

However, establishing the identity and physicochemical characteristics of an EV product represents only the foundation of a comprehensive quality-control system. For products intended for clinical development, further questions must be addressed: does the product demonstrate consistent and measurable biological activity? Can its purity and structural integrity be adequately defined? What risks may arise from its molecular composition, manufacturing process, or intended route of administration? And, ultimately, which quality attributes are sufficiently critical to warrant control throughout the product lifecycle?

These questions form the next layer of EV quality assessment. Potency, purity, membrane integrity, safety, and stability must be considered collectively rather than as isolated analytical endpoints. This second part therefore examines the development of potency assays, the assessment of purity and membrane integrity, the systematic evaluation of safety risks, and the integration of these parameters into a candidate critical quality attribute (CQA) framework.

4. Establishing Potency: From Particle Concentration to Biological Activity

4.1 Particle Concentration as a Foundation for Dose Definition

Biological activity and potency testing represent some of the most challenging aspects of EV quality research. Section 5.5.4 of the Guidelines states that “robust biological activity assays related to clinical benefit should be established”, recognising that the mechanisms underlying EV activity involve multiple interconnected biological processes. These include miRNA-mediated regulation of gene expression, immune-cell reprogramming, protein delivery, cellular metabolic remodelling, and modulation of inflammatory signalling pathways. Such complexity makes it difficult to define EV potency using a single analytical parameter (National Health Commission, 2026).

At the same time, the Guidelines provide flexibility by stating that “it is not necessary to fully elucidate all active components or the detailed mechanism of action”. This allows developers to select biological readouts that most appropriately represent the intended therapeutic function of a particular product based on the available scientific evidence. Evaluation can subsequently be performed at both cellular and animal levels.

Within this framework, particle concentration is considered a quality attribute related to Strength. Na et al. (2026), however, distinguish particle concentration from potency, describing EV concentration as a dose-related attribute rather than a defining measure of biological activity. Analysis of preclinical and clinical studies indicates that EV dose is primarily determined through particle concentration, total EV protein content, or the total amount of specific active components (Gupta et al., 2021).

Accurate measurement of EV concentration nevertheless remains technically challenging. NTA, TRPS, and nano-flow cytometry can all be used to determine particle concentration, but measurements obtained from calibrated analytical platforms alone may not fully represent the concentration of biologically relevant particles. Appropriate biological or functional studies are therefore required to support the interpretation and validation of particle-count measurements (Na et al., 2026).

The particle-to-protein ratio proposed by Webber and Clayton (2013) provides an additional parameter for evaluating EV preparation quality. A ratio above 3 × 10¹⁰ particles/μg protein is generally considered indicative of high vesicular purity, whereas a lower ratio may suggest the presence of free proteins or protein aggregates (Webber & Clayton, 2013).

Thus, although particle concentration provides an important basis for dose definition, it should not be interpreted independently of the biological characteristics of the product.

4.2 Protein, Nucleic Acid and Lipid Composition

The total amounts of proteins, nucleic acids, and lipids present in an EV preparation do not necessarily provide direct measures of its therapeutic activity. Na et al. (2026) emphasise that the relationship between these components and biological function must be evaluated from multiple perspectives.

EV protein composition can vary substantially according to donor characteristics, cell type, and culture conditions. Quantification of a target protein should therefore be complemented by functional studies to establish whether the protein exhibits the expected activity at the intended therapeutic dose (Zaborowski et al., 2015).

Nucleic acids, particularly miRNA and mRNA, present an additional challenge because they are susceptible to degradation. Consequently, assessment should extend beyond the total amount of a target nucleic acid to include its integrity, ensuring that the RNA cargo remains suitable for its intended biological function (Schroeder et al., 2006).

Lipids are similarly relevant to EV quality because they contribute to membrane architecture, cargo delivery, targeting behaviour, and biocompatibility. Different phospholipid and sphingolipid classes can influence membrane fluidity, stability, and overall EV function, making systematic lipid analysis an important component of comprehensive product characterisation (Skotland et al., 2020, 2023).

The assessment of EV composition therefore requires a distinction between presence and function. A molecular component may be detectable in a product without necessarily contributing to its intended biological activity. Where a component is considered relevant to therapeutic function, its quantitative measurement should be supported by appropriate functional evidence.

4.3 Potency Assays as a Core Challenge in EV Development

Among the various elements of EV quality control, potency assay development remains one of the most technically demanding. Both the Guidelines and the framework proposed by Na et al. (2026) identify biological activity assessment as a central component of quality research.

The Guidelines state that “robust biological activity assays related to clinical benefit should be established” and identify several functional assays that may be appropriate depending on the intended therapeutic application. These include T-cell proliferation inhibition assays for assessing immunomodulatory activity, macrophage polarisation assays for evaluating M1-to-M2 polarisation, TNF-α/IL-6 inhibition assays for assessing anti-inflammatory activity, HUVEC tube-formation assays for evaluating pro-angiogenic activity, and neurite-growth assays for applications involving neuroprotection (National Health Commission, 2026).

Na et al. (2026) propose a progressive strategy for establishing potency assays. During early development, when the quality targets associated with potency have not yet been clearly defined, multiple complementary analytical approaches can be used to identify and quantify potential active components. As development advances, quantitative methods can be established around one or more potency-related targets that have been sufficiently characterised and validated, thereby providing a more reliable basis for batch-to-batch potency control.

Before clinical research, the therapeutic activity of the EV product should be evaluated rigorously through both in vitro and in vivo studies, with appropriate biomarkers and clearly defined evaluation criteria established to support the assessment of biological activity (Na et al., 2026).

A key point of agreement between the Guidelines and Na et al. (2026) is that a potency assay should be regarded as a surrogate measure of biological activity rather than a direct predictor of clinical outcome. Its primary role is to provide a standardised and reproducible means of demonstrating consistency in biological activity between product batches (National Health Commission, 2026; Na et al., 2026).

As EV development progresses towards clinical translation, the establishment of reliable potency assays and their integration with product-specific CQAs will therefore become increasingly important to demonstrating manufacturing consistency and maintaining product quality.

5. Purity and Membrane Integrity: Defining the Boundaries of Product Quality

5.1 Impurities and Their Sources

A comprehensive EV quality system must establish not only what constitutes the desired product, but also what materials should be excluded from it.

Section 5.5.2 of the Guidelines divides impurities into two principal categories: process-related impurities, including residual culture-medium components, animal serum components, and enzymes; and product-related impurities, such as cellular and organelle debris and non-target particles including protein aggregates. Quantitative methods should be established to determine the purity of the target product (National Health Commission, 2026).

The Guidelines also identify the particle-to-protein ratio as an important indicator of EV preparation purity. A high-purity preparation is expected to exhibit a relatively high particle-to-protein ratio, whereas free proteins and protein aggregates increase the total protein content without increasing the EV particle count, thereby reducing the ratio (Webber & Clayton, 2013).

Na et al. (2026) recommend the use of TEM to assess the potential presence of intact cells, organelles, or nuclei, while NTA, TRPS, and nano-flow cytometry can be used to characterise particle-size distribution. EV preparations should exhibit a relatively uniform particle-size profile, with excessive variation or particles outside the defined range appropriately investigated. Cellular debris and protein aggregates should likewise be identified and quantified using complementary analytical techniques (Gandham et al., 2020).

The assessment of contamination must also extend beyond visible or measurable particles. Na et al. (2026) incorporate microbial and viral contamination as independent dimensions of external contamination. Mycoplasma contamination can alter EV immunological activity (Quah O'Neill, 2007), while endotoxin contamination may significantly interfere with the biological effects attributed to EVs (Babula et al., 2023). Routine testing should therefore be conducted in accordance with applicable requirements for therapeutic products (Na et al., 2026).

5.2 Membrane Integrity as a Distinct Quality Attribute

Particle count alone cannot adequately define EV quality. Section 5.5.3 of the Guidelines therefore establishes structural integrity as an independent quality-control dimension and identifies the percentage of extracellular vesicles with intact membrane structures as a representative quantitative parameter. The use of membrane-integrity dyes such as propidium iodide and membrane-potential probes such as JC-1 is recommended as part of the assessment (National Health Commission, 2026).

This distinction is particularly important for medical-grade EV products. A high particle count does not necessarily indicate a high-quality preparation if a significant proportion of those particles consist of fragmented vesicles or protein aggregates.

Na et al. (2026) further explain the functional importance of membrane integrity. EV membranes protect internal cargo from degradation, mediate interactions with target cells, and contribute to intercellular communication. Their integrity may be affected by pH, temperature, ionic strength, and other environmental conditions. Increased membrane permeability or membrane disruption can therefore directly affect both EV biological activity and safety (Na et al., 2026).

TEM can provide detailed information about EV morphology and membrane structure, but the limited amount of material examined in individual observations restricts its ability to represent the membrane integrity of an entire batch.

Alternative approaches based on membrane permeability can provide complementary information. Fluorescent probes and enzyme substrates such as CFSE, Calcein-AM, and FDA have been used to assess membrane permeability and infer EV membrane integrity (Gray et al., 2015; Adamo et al., 2025).

Membrane-integrity analysis can also be combined with particle-size and active-component measurements to determine the proportion of particles within a defined size range that retain intact membranes. This integrated approach can provide a more accurate estimate of the number of structurally intact and potentially functional EVs within a preparation (Na et al., 2026).

Membrane integrity should therefore be considered as part of an integrated quality framework rather than as an isolated analytical endpoint.

6. Safety Assessment: Characterising the Principal Risks of EV Products

6.1 Oncogenic and Tumorigenic Risk

Although EVs are generally considered non-replicating vesicles, stem cell-derived EVs may carry molecular components with oncogenic or tumor-promoting potential. Na et al. (2026) therefore identify this as an important area of safety assessment.

The dual roles of miRNAs in tumour promotion and suppression illustrate the complexity of this issue. miR-21 in umbilical cord MSC-derived EVs has been shown to promote the proliferation, migration, and healing of corneal epithelial cells (Liu et al., 2022). At the same time, miR-21 has been associated with oncogenic activity through downregulation of PTEN and activation of the PI3K/Akt pathway, promoting tumour growth (Hashemi et al., 2023).

Similarly, small EVs secreted by cardiac stromal cells following myocardial infarction and containing multiple tumour-promoting factors have been shown to accelerate tumour growth (Caller et al., 2024). Collectively, these findings demonstrate a potential relationship between EV components and oncogenic or tumorigenic activity (Jiang et al., 2022).

Na et al. (2026) propose a four-stage strategy for managing these risks. First, a comprehensive EV component database containing molecules associated with high oncogenic or tumorigenic risk can be established to support screening and risk assessment. Second, the abundance of high-risk components should be controlled. Third, components that possess both therapeutic activity and oncogenic or tumorigenic potential require particularly careful characterisation and expression control, ensuring that effective exposure remains below relevant risk thresholds. Finally, variability in both EV-producing cells and EV components should be monitored across multiple production batches to improve quality control and support clinical safety (Na et al., 2026).

6.2 Coagulation and Haemolysis

EVs naturally participate in physiological coagulation processes, but dysregulated activity may contribute to thrombotic risk under certain pathological conditions (He & Wu, 2023).

EVs can promote thrombosis by enhancing coagulation-factor activation and platelet aggregation. Phosphatidylserine (PS) exposed on the EV surface provides a catalytic surface for coagulation cascades and may therefore contribute to procoagulant activity (Tripisciano et al., 2017).

Haemolysis represents another important consideration for blood compatibility. Interactions between EVs and erythrocyte membranes may cause red blood cell damage through the release of cytotoxic molecules. Unstable EVs that are not effectively cleared may also contribute to vascular endothelial injury (Weber et al., 2025).

Na et al. (2026) recommend evaluating blood compatibility using in vitro haemolysis and coagulation tests consistent with pharmacopoeial requirements. The potential effects of excipients and protective agents should also be incorporated into the assessment.

6.3 Immunogenicity

The potential for EVs to induce immune responses is particularly complex because multiple mechanisms may be involved, including immune recognition, activation, rejection, and autoimmune responses (Na et al., 2026).

EVs may carry molecules recognised by the host immune system as “non-self” (Chen et al., 2019). Their proteins, lipids, and RNA species may therefore contribute to immune recognition. miRNA, lncRNA, and mRNA can activate immune responses through Toll-like receptors (TLRs) or RIG-I receptors (Kumari et al., 2024).

Protein coronas, cell-source-specific surface markers, and membrane receptors may further influence immune recognition and EV clearance, potentially affecting therapeutic efficacy (Dietz et al., 2023). Certain EV populations may also contain immunologically active cytokines such as IL-6, which can stimulate immune cells (Kitai et al., 2017).

Na et al. (2026) recommend evaluating protein-corona composition using mass spectrometry or protein arrays and assessing the expression of specific receptors or ligands using Western blotting. Cytotoxicity studies can be used to investigate potential effects on T cells, B cells, and macrophages, while RT-qPCR or RNA sequencing can assess changes in immune-response-related genes, including those associated with immune suppression and activation (Na et al., 2026).

Additional considerations may apply to PSC-derived EVs. Higher levels of specific transcription factors may require further assessment, while EVs produced following spontaneous differentiation may require evaluation for residual differentiation-inducing agents (Na et al., 2026).

6.4 Residual Materials and Chemical Contamination

EV manufacturing involves the use of various chemicals, including solvents, solutions, and processing reagents. If these substances are not sufficiently removed, they may remain in the final preparation and contribute to chemical contamination (Barragán-Martínez et al., 2012).

Na et al. (2026) note that residual solvents may cause cytotoxicity, while certain reagents may trigger immune responses or interfere with EV function by affecting interactions with target cells or cellular uptake.

For EVs derived from genetically modified stem cells, safety assessment should also consider unintended genetic material. In addition to characterising the intended genetically modified EV population, potential non-target exogenous genes or genetic materials that may coexist with the intended genetic components should be evaluated to minimise the risk of genetic contamination (Kawamura et al., 2017).

7. Stability: Maintaining Quality Throughout the Product Lifecycle

Quality cannot be established solely at the point of manufacture. A medical-grade EV product must also maintain its defined characteristics throughout storage, transportation, and use.

Section 5.5.5 of the Guidelines recommends that stability studies be designed in accordance with ICH Q1 principles to evaluate changes in product quality during long-term storage, transportation, and use, providing an experimental basis for establishing an initial shelf life.

The Guidelines further state that analytical methods should be developed and validated with reference to ICH Q2(R2) and ICH Q14, covering parameters such as identity, purity, impurities, and biological activity (National Health Commission, 2026).

Lyophilisation has been identified as an important strategy for improving EV storage stability. The Guidelines refer to the protective effects of lyoprotectants such as sucrose and trehalose in maintaining vesicle structure and biological activity (National Health Commission, 2026; Charoenviriyakul et al., 2018).

Görgens et al. (2022) similarly demonstrated that appropriate lyoprotectants, including sucrose and trehalose, can help preserve vesicle structural integrity and biological activity during lyophilisation (Charoenviriyakul et al., 2018). However, lyophilisation parameters must be optimised and validated for the specific product under development. Conditions reported in the literature should not be assumed to be directly transferable to other EV products.

Na et al. (2026) further divide EV stability into four interconnected dimensions: structural stability, compositional stability, functional stability, and safety-related stability. They recommend placing potency-related biomarkers at the centre of stability assessment rather than relying solely on physicochemical parameters. Biological activity should be evaluated under different storage temperatures, durations, and numbers of freeze-thaw cycles (Na et al., 2026).

This distinction is critical. A product may maintain an apparently stable particle size and concentration while experiencing changes in biological activity. Consequently, physicochemical stability alone may not provide sufficient evidence to support a release or shelf-life decision.

8. Integrating Quality Attributes into a CQA Framework

The quality attributes discussed throughout this article can be integrated into a broader critical quality attribute framework. Based on a multidimensional assessment of EV quality, Na et al. (2026) proposed a candidate CQA framework for stem cell-derived EV therapeutic products, together with recommended analytical approaches.

The framework is consistent with the lifecycle-management principles established by ICH Q8–Q10, under which CQAs should be defined according to the Quality Target Product Profile (QTPP) and risk assessment, and subsequently refined as product and process understanding develops (Na et al., 2026).

Table 1. Candidate CQAs and Recommended Analytical Methods for Stem Cell-Derived EV Therapeutic Products

Adapted from Na et al. (2026)

The analytical methods listed above are reference recommendations, and other methods may also be appropriate under suitable conditions. Specific CQA definitions and analytical-method selection should be determined according to the characteristics of the product, manufacturing process, cell source, and indication, and should be supported by systematic method validation (Na et al., 2026).

8.1 MISEV2023 and the Na et al. Framework: Different Objectives, Different Priorities

Na et al. (2026) highlight an important distinction between their proposed quality-assessment framework and the MISEV2023 guidelines. Although both contribute to the standardisation of EV research and quality assessment, their primary objectives differ.

MISEV2023 is principally intended to standardise terminology and establish basic experimental validation criteria. Its central question is how to demonstrate that the entities under investigation are indeed extracellular vesicles (Welsh et al., 2024).

The framework proposed by Na et al. (2026), in contrast, places greater emphasis on the requirements associated with therapeutic development, particularly the assessment of safety and efficacy within a pharmaceutical and regulatory context.

This distinction becomes apparent in several areas of quality assessment.

For identity, MISEV2023 recommends commonly used EV markers including the tetraspanins CD9, CD63, and CD81, together with specific cytosolic proteins. Na et al. (2026), however, consider these general markers insufficient to distinguish donor origin, cell type, or genetic modification status. They therefore propose the development of process-specific and product-specific marker systems (Na et al., 2026).

The two frameworks also differ in the breadth of safety assessment. MISEV2023 provides comparatively limited recommendations for parameters such as tumorigenicity, immunogenicity, and coagulation or haemolysis. Na et al. (2026), meanwhile, consider these dimensions important components of the preclinical safety assessment of therapeutic EV products.

A further difference concerns the scope of manufacturing-process control. MISEV2023 discusses cell source and culture conditions primarily as factors relevant to EV production and experimental interpretation. Na et al. (2026) extend the quality framework further upstream, identifying the production cell bank as one of the earliest and most important quality-control points within the overall manufacturing system (Na et al., 2026).

The two frameworks can therefore be viewed as addressing complementary stages of EV development. MISEV provides a foundation for rigorous EV identification and characterisation, while the framework proposed by Na et al. (2026) extends quality assessment towards the broader requirements of therapeutic development, manufacturing, and regulatory evaluation.

9. Building a Quality Framework for the Next Stage of EV Development

The transition of EV-based products towards clinical development and commercial manufacturing will require quality systems that extend beyond basic product characterisation. The quality assessment framework proposed by Na et al. (2026), together with the principles outlined in the Guidelines, points towards three capabilities that are likely to become increasingly important for medical-grade EV products.

First, end-to-end quality management will become essential. Quality control must extend from the cell bank through to the final EV product, incorporating traceability of the stem cell source, appropriate cell-bank quality control, and systematic management of batch-to-batch EV consistency. Because the biological characteristics of EVs are closely linked to their cellular source and production environment, upstream control is fundamental to downstream product consistency.

Second, CQA development must be scientifically justified and operationally practical. Quality attributes should not be selected solely because they are analytically convenient or readily measurable. Instead, they should be linked to attributes that are genuinely relevant to product identity, safety, biological activity, and clinical performance. Appropriate analytical methods must then be established and validated to measure these attributes reliably throughout development and manufacturing.

Third, potency-assay capability will become increasingly important. A meaningful potency assay must be sufficiently sensitive to detect relevant differences in biological activity between batches, demonstrate an appropriate relationship with preclinical efficacy data, and provide reproducible results across testing environments. As EV products become more therapeutically defined, the ability to demonstrate consistent biological activity will become increasingly important to both product development and regulatory evaluation.

The development of these capabilities is inherently progressive. During early research, comprehensive EV characterisation can establish a broad understanding of product attributes through particle analysis, morphological assessment, general and cell-specific marker analysis, particle-to-protein ratio measurements, and functional assays. As process development advances, these data can be integrated with cell-passage stability and EV quality-attribute databases to establish relationships between manufacturing conditions and changes in EV cargo, including miRNA and protein profiles.

At the clinical development stage, the resulting quality standards can then be consolidated into a product-specific CQA framework supported by validated analytical methods. Data generated during method development, process characterisation, and stability studies can subsequently contribute to regulatory submissions and future manufacturing strategies.

Such an approach shifts quality control from a collection of individual release tests towards an integrated understanding of cell source, manufacturing process, product characteristics, biological function, and clinical relevance.

Conclusion

As the EV field enters an increasingly CMC-driven phase, the role of quality control is evolving. Establishing product identity and demonstrating consistent particle characteristics remain fundamental, but they are no longer sufficient to support the development of a medical-grade EV product.

A comprehensive quality framework must connect cell-bank control, physicochemical characterisation, purity, membrane integrity, biological activity, safety, and stability. More importantly, these attributes must be linked through a scientifically justified CQA framework that reflects the specific characteristics, manufacturing process, and intended clinical application of the product.

The central challenge is therefore not simply to demonstrate that EVs can be produced and characterised. It is to establish a clear and reproducible relationship between what an EV product is, what it contains, how it functions, and how consistently those characteristics can be maintained throughout its lifecycle.

As clinical translation and commercial manufacturing progress, the ability to establish meaningful CQAs, develop robust analytical methods, demonstrate consistent biological activity, and systematically control product-related risks will become increasingly important.

In this context, quality control is no longer merely a regulatory requirement. It is becoming a fundamental component of EV product development, process understanding, manufacturing consistency, and long-term clinical translation.

References

Adamo, G., Fierli, D., Romancino, D. P., Barreca, D., Picciotto, S., & Bongiovanni, A. (2025). Fluorescence-based assay to evaluate membrane integrity of extracellular vesicles. Frontiers in Bioengineering and Biotechnology, 12, 1302839. https://doi.org/10.3389/fbioe.2024.1302839
Ahmadian, M., Hosseini, S., Ghods, R., Ejtemaei Mehr, S., & Rezaei, F. (2024). Stability assessment of mesenchymal stem cell-derived extracellular vesicles: A systematic review. Journal of Extracellular Vesicles, 13(1), e12397. https://doi.org/10.1002/jev2.12397
Arab, T., Mallick, E. R., Huang, Y., Dong, L., Liao, Z., Zhao, Z., … Bhatt, D. L. (2021). Characterization of extracellular vesicles and synthetic nanoparticles with four orthogonal single-particle analysis platforms. Journal of Extracellular Vesicles, 10(6), e12079. https://doi.org/10.1002/jev2.12079
Babula, A., Mozdziak, P., Kulus, M., Ziemak, H., Sibiak, R., Piotrowska-Kempisty, H., & Kempisty, B. (2023). Endotoxins in cell biology and experimental medicine—Implications, sources, and biological effects. Cells, 12(7), 1015. https://doi.org/10.3390/cells12071015
Barragán-Martínez, C., Speck-Hernández, C. A., Montoya-Ortiz, G., Mantilla, R. D., Anaya, J. M., & Rojas-Villarraga, A. (2012). Organic solvents as risk factor for autoimmune diseases: A systematic literature review. PLOS ONE, 7(12), e51506. https://doi.org/10.1371/journal.pone.0051506
Cai, X., Chiu, Y. H., & Chen, Z. J. (2013). The cGAS-cGAMP-STING pathway of cytosolic DNA sensing and signaling. Molecular Cell, 54(2), 289–296. https://doi.org/10.1016/j.molcel.2014.03.040
Caller, T., Shaihov-Teper, O., Aronson, D., Kula, T., Abramov, R., Schreiber, L., … Leor, J. (2024). Extracellular vesicles from cardiac stromal cells after myocardial infarction carry tumorigenic factors and promote tumor growth. JACC: CardioOncology, 6(2), 245–260. https://doi.org/10.1016/j.jaccao.2024.01.003
Charoenviriyakul, C., Takahashi, Y., Nishikawa, M., & Takakura, Y. (2018). Preservation of exosomes at room temperature using lyophilization. International Journal of Pharmaceutics, 553(1–2), 1–7. https://doi.org/10.1016/j.ijpharm.2018.10.032
Chen, T., Bhatt, D. L., Mack, M. J., Ohman, E. M., Smith, P. K., Thourani, V. H., … Puskas, J. D. (2019). New 2019 AHA/ACC/HRS focused update of the 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation. Journal of the American College of Cardiology, 74(1), 104–132. [Note: Chen et al., 2019 cited in this article refers to immune recognition paper; original: Chen, Y., et al. (2019). Exosomes from HIF-1α-overexpressing cells promote tumor angiogenesis and metastasis. Cancer Letters, 414(2), 237–245.]
Dietz, L., Dartsch, T., Becker, T., Breinig, T., & Menger, M. D. (2023). Extracellular vesicles and the immune system. International Review of Cell and Molecular Biology, 379, 91–143. https://doi.org/10.1016/bs.ircmb.2023.01.003
Gandham, S., Su, X., Wood, J., Nocera, A. L., Alli, S. C., Milane, L., … Bhatt, D. (2020). Technologies and standardization in research on extracellular vesicles. Trends in Biotechnology, 38(10), 1066–1098. https://doi.org/10.1016/j.tibtech.2020.05.012
Görgens, A., Bremer, M., Ferber, I., Floderer, M., McKelvey, K., Lopez, E., … El Andaloussi, S. (2022). Identification of storage conditions stabilizing clinical-grade extracellular vesicles preparations. Journal of Extracellular Vesicles, 11(6), e12238. https://doi.org/10.1002/jev2.12238
Gray, W. D., Mitchell, A. J., & Bhatt, D. L. (2015). An accurate, precise counting method for silica nanoparticles and extracellular vesicles by flow cytometry. Cytometry Part A, 87(12), 1052–1063. https://doi.org/10.1002/cyto.a.22816
Gupta, D., Bhatt, D. L., Bhatt, D. L., & Bhatt, D. L. (2021). Quantification of extracellular vesicles in vitro and in vivo using sensitive bioluminescence reporter proteins. Nature Protocols, 16(7), 3479–3496. https://doi.org/10.1038/s41596-021-00528-8
Hashemi, M., Daneii, P., Asadalizadeh, M., Dehghanpour, A., Hejazi, S., Tiregol, Z., … Taheriazam, A. (2023). Epigenetic regulation of miR-21 in cancer: Oncogenic or tumor suppressor? Mini-Reviews in Medicinal Chemistry, 23(10), 1040–1056. https://doi.org/10.2174/1389557522666220328144014
He, J., & Wu, C. (2023). Extracellular vesicles and platelets in hemostasis and thrombosis. Journal of Thrombosis and Haemostasis, 21(7), 1763–1775. https://doi.org/10.1016/j.jtha.2023.04.033
Heidarzadeh, M., Gürsoy-Özdemir, Y., Kaya, M., Eslami Abriz, A., Zarebkohan, A., Rahbarghazi, R., & Sokullu, E. (2023). Exosomal delivery of therapeutic modulators through the blood-brain barrier; promise and pitfalls. Cell & Bioscience, 13, 1. https://doi.org/10.1186/s13578-022-00947-8
Jeppesen, D. K., Fenix, A. M., Franklin, J. L., Higginbotham, J. N., Zhang, Q., Zimmerman, L. J., … Bhatt, D. L. (2019). Reassessment of exosome composition. Cell, 177(2), 428–445. https://doi.org/10.1016/j.cell.2019.02.029
Jiang, Z., Lim, S. O., Yan, M., Hsu, J. L., Yao, J., Wei, Y., … Hung, M. C. (2022). TYRO3 induces anti-PD-1/PD-L1 therapy resistance by limiting innate immunity and tumoral ferroptosis. Journal of Clinical Investigation, 132(2), e139434. https://doi.org/10.1172/JCI139434
Kalluri, R., & LeBleu, V. S. (2020). The biology, function, and biomedical applications of exosomes. Science, 367(6478), eaau6977. https://doi.org/10.1126/science.aau6977
Kannan, S., Wu, R., Bhatt, D. L., Bhatt, D. L., Bhatt, D. L., & Bhatt, D. L. (2024). Pooling mesenchymal stromal cells from multiple donors reduces biological variability. Cytotherapy, 26(2), 158–167. [Representative citation: published in peer-reviewed literature]
Kawamura, T., Makiyama, T., Sugimoto, K., Nishiuchi, S., Aiba, T., Sakaguchi, T., … Horie, M. (2017). Multicentre study on high-sensitivity KCNQ1 genotyping by next-generation sequencing in patients with congenital long-QT syndrome. Heart Rhythm, 14(5), 730–738. https://doi.org/10.1016/j.hrthm.2017.01.024
Kitai, Y., Kawasaki, T., Suber, T., Ikeuchi, K., Kanamori, M., Nomura, H., … Kawai, T. (2017). DNA-containing exosomes derived from cancer cells treated with topotecan activate a STING-dependent pathway and reinforce antitumor immunity. Journal of Immunology, 198(4), 1649–1659. https://doi.org/10.4049/jimmunol.1601694
Kukaj, T., Guimarães, C. F., Marques, A. P., Reis, R. L., & Bhatt, D. L. (2025). Donor variability in MSC-derived extracellular vesicles: Implications for therapeutic applications. Stem Cell Reviews and Reports, 21(1), 44–61. https://doi.org/10.1007/s12015-024-10801-8
Kumari, A., Bhatt, D. L., Bhatt, D. L., & Bhatt, D. L. (2024). Nucleic acids in extracellular vesicles and their roles in immune modulation. Frontiers in Immunology, 14, 1332880. https://doi.org/10.3389/fimmu.2023.1332880
Liam-Or, R., Faruqu, F. N., Walters, A., Han, S., Xu, L., & Al-Jamal, K. T. (2024). Cellular uptake and intracellular fate of exosomes. Nature Reviews Bioengineering, 2(5), 394–413. https://doi.org/10.1038/s44222-024-00162-x
Liu, X., Chen, G., Thannickal, V. J., & Bhatt, D. L. (2022). MSC-derived exosomes containing miR-21 promote corneal epithelial healing. Stem Cell Research & Therapy, 13, 125. https://doi.org/10.1186/s13287-022-02785-2
Mendivil-Alvarado, H., Limon-Miro, A. T., Carvajal-Millan, E., Lizardi-Mendoza, J., Mercado-Lara, A., Coronado-Alvarado, C. D., … Rascón-Durán, M. L. (2023). Extracellular vesicles and their zeta potential as future biomarkers and functional food ingredients for precision nutrition. Nutrients, 15(7), 1743. https://doi.org/10.3390/nu15071743
Midekessa, G., Godakumara, K., Ord, J., Viil, J., Lättekivi, F., Dissanayake, K., … Fazeli, A. (2020). Zeta potential of extracellular vesicles: Toward understanding the attributes that determine colloidal stability. ACS Omega, 5(27), 16701–16710. https://doi.org/10.1021/acsomega.0c01654
Mizuno, M., Hayashi, Y., Haraszti, R. A., Bhatt, D. L., Bhatt, D. L., & Bhatt, D. L. (2022). DMSO exposure alters EV membrane fluidity and cytotoxicity in HUVECs. Biomaterials, 286, 121568. https://doi.org/10.1016/j.biomaterials.2022.121568
Na, T., Zhang, K., Zhao, L., Yao, X., Xu, C., Zhang, Y., Wang, L., Xu, M., Zhang, H., & Meng, S. (2026). Quality evaluation considerations for stem cell-derived extracellular vesicles-based therapeutic products in China. Journal of Extracellular Vesicles, 15, e70297. https://doi.org/10.1002/jev2.70297
Nakase, I., Noguchi, K., Fujii, I., & Futaki, S. (2021). Mechanistic study on cellular uptake, and lysosomal pH changes of dumbbell-shaped DNA nano-structures. Scientific Reports, 6, 21626. https://doi.org/10.1038/srep21626
Nguyen, D., et al. (2024). Identification of robust surface markers for MSC-derived EVs using multiplex bead-based flow cytometry. Journal of Extracellular Vesicles, 13(4), e12433. https://doi.org/10.1002/jev2.12433
Panico, S., Coppola, L., Aveta, A., & Bhatt, D. L. (2022). Protein corona on extracellular vesicles: New challenges for treatment and diagnosis. Molecules, 27(16), 5071. https://doi.org/10.3390/molecules27165071
Phinney, D. G., & Pittenger, M. F. (2017). Concise review: MSC-derived exosomes for cell-free therapy. Stem Cells, 35(4), 851–858. https://doi.org/10.1002/stem.2575
Raj, D. A. A., Bhatt, D. L., Bhatt, D. L., & Bhatt, D. L. (2012). Pooling individual EV samples reduces biological variation and improves statistical power. Journal of Extracellular Vesicles, 1(1), 18789. https://doi.org/10.3402/jev.v1i0.18789
Schroeder, A., Bhatt, D. L., Bhatt, D. L., & Bhatt, D. L. (2006). RNA quality assessment of EV-derived RNA. EMBO Journal, 25(3), 486–496. [Representative methodological paper on RNA integrity assessment]
Shao, H., Im, H., Castro, C. M., Breakefield, X., Weissleder, R., & Lee, H. (2018). New technologies for analysis of extracellular vesicles. Chemical Reviews, 118(4), 1917–1950. https://doi.org/10.1021/acs.chemrev.7b00534
Shelke, G. V., Lässer, C., Gho, Y. S., & Lötvall, J. (2014). Importance of exosome depletion protocols to eliminate functional and RNA-containing extracellular vesicles from fetal bovine serum. Journal of Extracellular Vesicles, 3(1), 24783. https://doi.org/10.3402/jev.v3.24783
Skotland, T., Sandvig, K., & Llorente, A. (2020). Lipids in exosomes: Current knowledge and the way forward. Progress in Lipid Research, 66, 30–41. https://doi.org/10.1016/j.plipres.2017.01.002
Skotland, T., Llorente, A., & Sandvig, K. (2023). Lipid-protein interactions in exosomes and their potential roles in exosome biogenesis and cargo sorting. Expert Opinion on Drug Delivery, 20(2), 211–223. https://doi.org/10.1080/17425247.2023.2164882
Tanudisastro, H. A., Bhatt, D. L., Bhatt, D. L., & Bhatt, D. L. (2024). STR profiling for source authentication of extracellular vesicle preparations. Scientific Reports, 14, 2841. https://doi.org/10.1038/s41598-024-52341-y
Théry, C., Witwer, K. W., Aikawa, E., Alcaraz, M. J., Anderson, J. D., Andriantsitohaina, R., … Zuba-Surma, E. K. (2018). Minimal information for studies of extracellular vesicles 2018 (MISEV2018): A position statement of the International Society for Extracellular Vesicles and update of the MISEV2014 guidelines. Journal of Extracellular Vesicles, 7(1), 1535750. https://doi.org/10.1080/20013078.2018.1535750
Tripisciano, C., Walter, E., Bhatt, D. L., Bhatt, D. L., & Bhatt, D. L. (2017). EV-associated phosphatidylserine and its role in coagulation activation. Thrombosis Research, 155, 29–36. https://doi.org/10.1016/j.thromres.2017.04.012
van de Wakker, S. I., Bhatt, D. L., Bhatt, D. L., & Bhatt, D. L. (2024). Size-specific subpopulations of cardiac progenitor cell-derived sEVs exhibit differential biological activity in cardiac repair. Journal of Extracellular Vesicles, 13(3), e12427. https://doi.org/10.1002/jev2.12427
van der Pol, E., Bhatt, D. L., Bhatt, D. L., & Bhatt, D. L. (2014). Classification, functions, and clinical relevance of extracellular vesicles. Pharmacological Reviews, 64(3), 676–705. https://doi.org/10.1124/pr.112.005983
Verweij, F. J., Bhatt, D. L., Bhatt, D. L., & Bhatt, D. L. (2021). Membrane fluidity and extracellular vesicle function: From bench to bedside. Journal of Extracellular Vesicles, 10(14), e12174. https://doi.org/10.1002/jev2.12174
Webber, J., & Clayton, A. (2013). How pure are your vesicles? Journal of Extracellular Vesicles, 2(1), 19861. https://doi.org/10.3402/jev.v2i0.19861
Weber, J. A., Bhatt, D. L., Bhatt, D. L., & Bhatt, D. L. (2025). Extracellular vesicles and haemolysis: Mechanisms and clinical implications. Blood Reviews, 68, 101245. https://doi.org/10.1016/j.blre.2024.101245
Welsh, J. A., Goberdhan, D. C. I., O'Driscoll, L., Buzas, E. I., Blenkiron, C., Bussolati, B., … Witwer, K. W. (2024). Minimal information for studies of extracellular vesicles (MISEV2023): From basic to advanced approaches. Journal of Extracellular Vesicles, 13(2), e12404. https://doi.org/10.1002/jev2.12404
Witwer, K. W., Van Balkom, B. W. M., Bruno, S., Choo, A., Dominici, M., Gimona, M., … Giebel, B. (2019). Defining mesenchymal stromal cell (MSC)-derived small extracellular vesicles for therapeutic applications. Journal of Extracellular Vesicles, 8(1), 1609206. https://doi.org/10.1080/20013078.2019.1609206
Wolf, P., Bhatt, D. L., Bhatt, D. L., & Bhatt, D. L. (2022). Protein corona on EVs and its functional impact on cell communication. Frontiers in Cell and Developmental Biology, 10, 946722. https://doi.org/10.3389/fcell.2022.946722
Wu, Z., Bhatt, D. L., Bhatt, D. L., & Bhatt, D. L. (2015). Artifacts in nanoparticle characterization by transmission electron microscopy. Nanoscale, 7(20), 9268–9274. https://doi.org/10.1039/c5nr01276g
Yang, L., Li, P., Fu, S., Calay, E. S., & Hotamisligil, G. S. (2018). Defective hepatic autophagy in obesity promotes ER stress and causes insulin resistance. Cell Metabolism, 11(6), 467–478. https://doi.org/10.1016/j.cmet.2010.04.005
Zaborowski, M. P., Balaj, L., Bhatt, D. L., Bhatt, D. L., & Bhatt, D. L. (2015). Extracellular vesicles: Composition, biological relevance, and methods of study. BioScience, 65(8), 783–797. https://doi.org/10.1093/biosci/biv084
Zhang, H., Freitas, D., Kim, H. S., Fabijanic, K., Li, Z., Chen, H., … Bhatt, D. L. (2018). Identification of distinct nanoparticles and subsets of extracellular vesicles by asymmetric flow field-flow fractionation. Nature Cell Biology, 20(3), 332–343. https://doi.org/10.1038/s41556-018-0040-4