Skip to main content
NCBI home page
Journal of Extracellular Vesicles logoLink to Journal of Extracellular Vesicles
. 2026 Jun 30;15(7):e70330. doi: 10.1002/jev2.70330

Integrating Nuclear Magnetic Resonance into Bioprocess Control of Clinical‐Grade Extracellular Vesicle Manufacturing

Santiago Roura 1,2,3,4,✉, Núria Amigó 5,✉, Joaquim Vives 6,7,8,✉
PMCID: PMC13317633  PMID: 42377990

ABSTRACT

The rapid advance in the research and development of extracellular vesicle (EV)‐based therapeutics has stimulated a paradigm shift in the field of regenerative medicine. However, translating EV‐based therapies into the clinic requires robust, scalable, and Good Manufacturing Practice (GMP)‐compliant bioprocesses that ensure product consistency, potency and safety. In this Perspective, we propose that metabolomics, particularly by using high‐resolution nuclear magnetic resonance (NMR), can serve as a transformative analytical technology in EV manufacturing and quality control. By integrating NMR‐metabolomics monitoring into upstream cell culture and downstream EV purification workflows, it becomes possible to identify metabolic fingerprints predictive of cell performance, EV yield and EV bioactivity. Drawing from the experience of the GALVANO consortium, which is developing Spain's first GMP‐grade platform for EV manufacturing from clinical‐grade human Wharton Jelly's mesenchymal stromal cells (WJ‐MSC‐EVs) with particular promise in the modulation of inflammation and tissue regeneration, we highlight how NMR‐metabolomics can support Quality by Design (QbD) principles, enhance in‐process analytics and accelerate regulatory harmonisation. We further discuss the need for collaborative standardisation of analytical methods and reporting frameworks to ensure reproducibility and comparability across EV batches. Together, these strategies can advance EV‐based therapeutics toward reliable, large‐scale clinical application.

Keywords: bioprocess development, clinical grade, extracellular vesicles, Good Manufacturing Practice, mesenchymal stromal cells, NMR‐metabolomics, Quality by Design, translational medicine

Nuclear magnetic resonance (NMR) in bioprocess design for the production of extracellular vesicles (EVs). NMR‐metabolomics enables real‐time bioprocess intelligence, bridging biological complexity with GMP rigor and accelerating the clinical translation of EV‐based therapies. CPP, critical process parameters; CQA, Critical Quality Attributes; PAT, Process Analytical Technologies; QbD, Quality by Design; QC, quality control; QTTP, Quality Target Product Profile.

graphic file with name JEV2-15-e70330-g001.jpg

1. Introduction

Extracellular vesicles (EVs) are non‐replicative, lipid bilayer‐encapsulated nanoparticles that carry a diverse cargo of bioactive molecules (including proteins, lipids, glycans, nucleic acids, miRNAs), modulating intercellular communication and cell behaviour (Yáñez‐Mó et al. 2015). Thus, for regenerative purposes, EVs reproduce the state and therapeutic potential of their parental cells (that is, immunomodulation, angiogenesis and anti‐apoptotic signalling) while avoiding safety risks associated with live‐cell therapies, such as embolism, tumorigenicity and immune rejection (Li et al. 2022; Wu et al. 2020; Jiang et al. 2023; Zhu et al. 2024). However, to guarantee transition from bench to bedside, EV‐based therapeutic agents must meet the rigorous quality, purity, safety and efficacy standards of biological medicinal products. The central challenge remains the development of scalable, Good Manufacturing Practice (GMP)‐compliant and standardised bioprocesses capable of producing highly pure bona fide EVs with the most consistent potency and defined Critical Quality Attributes (CQAs) (Munoz‐Dominguez et al. 2021; Roura et al. 2017). Uncontrolled heterogeneity in vesicle subpopulations, cargo composition and particle‐to‐protein ratios can profoundly affect biological activity and compromise product comparability, underscoring the need for robust in‐process controls and orthogonal characterisation strategies. To this end, advanced Process Analytical Technologies (PAT) such as real‐time nanoparticle tracking, microfluidic flow cytometry and multi‐omics fingerprinting are increasingly recognised as essential to identify and assess the CQAs of EVs throughout upstream and downstream processing (Alexandre et al. 2023). Specifically, addressing these concerns requires the development of novel analytical or monitoring approaches, together with the adaptation and expanded use of existing tools within the EV manufacturing process.

In particular, nuclear magnetic resonance (NMR) spectroscopy provides rapid, non‐destructive, quantitative and structurally rich readouts of lipids, metabolites and nanoparticle‐associated signatures that have already been shown to discriminate and characterise EV preparations, thus representing an exceptionally well‐suited PAT for EV manufacturing and a powerful enabler of Quality‐by‐Design (QbD) strategies to ensure robust, traceable and clinically meaningful product quality (Lipsitz et al. 2016). Building upon this framework, this perspective provides a targeted synthesis of the literature at the intersection of NMR spectroscopy, MSC‐EV bioprocessing and GMP‐scale manufacturing, prioritising studies with high translational relevance to bridge the gap between analytical characterisation and clinical‐grade production.

2. Challenges in Bioprocessing Clinical‐Grade EVs

The current landscape of EV manufacturing is characterised by diversity in production methods and analytical standards, resulting in significant heterogeneity in EV preparations across laboratories and studies (Van Deun et al. 2017; Théry et al. 2018, Ter‐Ovanesyan et al., 2021), and therefore making comparability difficult. Key concerns relate to: (i) upstream process variability; (ii) downstream isolation and purification; (iii) analytical and potency testing; and (iv) regulatory alignment and quality control (QC), as briefly described next. Collectively, these challenges highlight the urgent need for integrated bioprocess analytical tools capable of capturing cellular and extracellular metabolic dynamics as surrogates for process quality, efficiency and product potency.

2.1. Upstream Process Variability

Most EV production still relies on two‐dimensional (2D) cell cultures using serum‐supplemented media, which introduces batch‐to‐batch variability and potential contamination with bovine‐derived EVs. Transitioning to chemically defined, xeno‐free media in three‐dimensional (3D) stirred‐tank bioreactors (STRs) allows improved scalability and control but introduces new critical process parameters (CPPs; e.g. microcarrier dynamics, shear stress, oxygenation) that must be tightly regulated (Lopez‐Fernandez et al. 2024).

2.2. Downstream Isolation and Purification

As mentioned above, the absence of universally accepted and standardised protocols for EV purification, such as differential ultracentrifugation, Tangential Flow Filtration (TFF) and size‐exclusion chromatography (SEC), may introduce a substantial bias in comparing EV yield, purity, and biological activity across GMP‐grade EV production platform for clinical applications. Ultracentrifugation, the most commonly used method, often leads to co‐isolation of protein aggregates and lipoproteins, as well as potential vesicle deformation due to high g‐forces (Clos‐Sansalvador et al. 2022; Coumans et al. 2017). In contrast, TFF enables scalable concentration but may alter vesicle surface properties through shear stress and membrane interactions (Busatto et al. 2018). On the other hand, SEC offers improved purity and preservation of EV integrity, yet at the expense of reduced recovery and scalability challenges (Böing et al. 2014; Gámez‐Valero et al. 2016). These methodological discrepancies hinder direct comparison of experimental outcomes and complicate the translation of EV‐based products into clinical‐grade therapeutics (Witwer et al. 2021; Lener et al. 2015). Moreover, the absence of harmonised QC criteria (regarding size distribution, surface markers and functional potency) further amplifies batch‐to‐batch variability (Van Deun et al. 2017; Théry et al. 2018). Therefore, establishing fit‐for‐purpose purification pipelines that balance efficiency, vesicle integrity and reproducibility, while complying with GMP standards, remains a critical bottleneck in the field (Chen et al. 2021). Developing such standardised and scalable workflows is essential to ensure their safety, efficacy and regulatory acceptance as new therapies. To this end, the pan‐European initiative driven by COST Action CA24114 (BTCs4ATMP: European network of SoHO establishments for rapid and sustainable access to ATMPs) represents a critical step toward establishing harmonised, GMP‐compliant scale‐up platforms for parental cell expansion, a necessary precondition for reproducible, large‐scale EV manufacture, thus providing the infrastructure and consensus needed to implement robust QC, batch‐to‐batch consistency and regulatory‐compliant EV products fit for clinical translation (www.cost.eu/actions/CA24114).

2.3. Analytical and Potency Testing

There is currently no universally accepted potency assay for EV‐based therapeutics. Similar to the analysis of potency in parental mesenchymal stromal cell (MSC) preparations, common surrogate assays for MSC‐EVs (such as T‐cell inhibition or CD73 enzymatic activity) provide partial insights but may not fully capture the complexity of EV bioactivity (Bauer et al. 2023). Furthermore, existing characterisation relies heavily on nanoparticle tracking analysis (NTA), flow cytometry and proteomics, which do not provide real‐time information during production. For functionalised EVs, potency metrics may be based on the detection and/or functionality of transgene or exogenous molecule introduced.

2.4. Regulatory Alignment and Quality Control

Regulatory frameworks for EV‐based products are evolving, but lack of harmonised definitions and analytical reference standards complicates clinical translation. Regulatory agencies and societies such as the International Society for Extracellular Vesicles (ISEV) and the International Society for Cell & Gene Therapy (ISCT) are moving toward consensus on identity, purity and potency requirements (Théry et al. 2018; Witwer et al. 2021; Börger et al. 2020), but alignment with GMP implementation remains incomplete.

3. NMR‐Metabolomics: A New Analytical Paradigm for EV Bioprocessing

Metabolomics provides a comprehensive snapshot of the biochemical state of cells and culture environments. By analysing small‐molecule metabolites (the end products of gene and protein activity), metabolomics enables direct insight into cellular function, viability and responses to stress (Sheng et al. 2025). To date, the complexity of EV preparations makes their purity assessment a demanding task. Fast and robust methods for EV characterisation are mandatory. In this context, NMR stands out as a high‐performance, high‐throughput platform with minimal analytical variability, based on the electromagnetic response to molecular excitation. As a spectroscopic technique, NMR provides spectral information about the chemical composition and physical status of analytes (Zhu and Raftery 2025). The main general advantages of the NMR‐based methods are: (i) no destruction of the sample; (ii) high analytical reproducibility; (iii) easy identification of molecular moieties; (iv) high robustness of instruments; (v) possibility to obtain molecular dynamics information; and, most importantly in this context (vi) direct quantitative information. With these advantages, the NMR‐based methods have the potential to offer the following specific valuable pros in EV bioprocess monitoring: (i) quantitative reproducibility: NMR produces inherently quantitative data without the need for internal calibration; (ii) minimal sample preparation: NMR can subsequently analyse conditioned media, bioreactor supernatants and EV suspensions with minimal processing; (iii) non‐destructive analysis: samples can be recovered for subsequent assays or reanalysis, (iv) High‐throughput compatibility: automated 1H‐NMR pipelines allow real‐time process monitoring in industrial settings.

In GMP settings, NMR metabolomics should currently be classified as an off‐line Process Analytical Technology (PAT) tool because it is most feasibly implemented off‐line, comprising aseptic sampling from closed bioreactor systems and final EV batches followed by analysis in dedicated facilities. Importantly, sterile single‐use or septum‐based sampling systems allow sample withdrawal without compromising system integrity or increasing contamination risks. Nevertheless, ongoing advances in compact and benchtop NMR instrumentation, together with increased automation and reduced costs, are expected to reach at‐line configurations in close proximity to production suites in the near future. In this context, recent perspectives have critically evaluated the lipid quantification methods recommended in MISEV guidelines, including fluorescent intercalating dyes, FTIR spectroscopy and the sulfo‐phospho‐vanillin assay, highlighting important limitations in their ability to accurately quantify total lipid content in EV preparations (Skotland et al. 2025). These limitations include dependence on membrane properties rather than lipid quantity, lack of demonstrated quantitative capability, and bias toward specific lipid classes such as unsaturated lipids. While mass spectrometry‐based lipidomics is currently considered the most comprehensive analytical approach, it typically requires complex sample preparation and is not readily adaptable to real‐time or process‐integrated monitoring. In contrast, NMR spectroscopy offers a complementary, non‐destructive and highly reproducible platform capable of providing quantitative and structurally informative lipid profiles with minimal sample preparation. Therefore, NMR could represent a valuable orthogonal approach to address some of the current analytical gaps in EV lipid characterisation.

3.1. Monitoring MSC Metabolism in Bioreactors

In the upstream phase, metabolomics can track cellular nutrient consumption, waste accumulation and metabolic fluxes, identifying “metabolic fingerprints” associated with high EV yield and desired potency. For instance, alterations in glycolytic and amino acid metabolism correlate with MSC senescence, while lipid biosynthesis pathways influence EV membrane composition and cargo loading (Wang et al. 2024; Zhang et al. 2025). NMR spectroscopy does not identify EVs through a single unique molecular marker, but rather through composite spectral patterns associated with lipid environments characteristic of membrane‐bound nanoparticles. In particular, EVs exhibit a phospholipid‐rich bilayer signature reflected in resonances from choline‐containing phospholipids and lipid acyl chains. Signals in regions of the 1H‐NMR spectrum (e.g. around ∼1.6 ppm) correspond to methylene groups in lipid acyl chains associated with membrane environments and contribute to the detection of intrinsic EV phospholipid content. We acknowledge that these signals are not inherently specific to EVs and may overlap with other lipid‐containing particles, such as lipoproteins or lipid nanoparticles. However, in our manufacturing framework, the use of chemically defined, serum‐free media significantly reduces the presence of exogenous lipoproteins. Under these conditions, the dominant phospholipid signals detected are expected to primarily originate from cell‐derived structures, including EVs.

3.2. Assessing EV Integrity and Composition

EVs carry distinguishable molecular cargo, including lipids, proteins, RNA and DNA, from their cell of origin. NMR‐based metabolomics, alone or in combination with other determinations, can potentially be applied to monitor the EV production and quantify particle concentration (Gandham et al. 2020). In particular, NMR can detect lipid profiles associated with EVs and other molecular signatures in EV preparations, serving as an additional QC layer for EV identity and batch‐to‐batch homogeneity. Indeed, particles to proteins (Pa/Pr) ratios determined by NTA, which permits the determination of both the size distribution and relative concentration of EVs, together with standardised BCA assay and the spectroscopic lipids to proteins (Li/Pr) ratio combined, may be reliable parameters to assess the EV purity with high precision and reproducibility. Among the characterisation techniques, the NMR spectroscopy presents some advantages over other metabolomic approaches based on mass spectrometry, since NMR spectra are obtained from intact plasma or serum matrices without prior extraction and can be calibrated to determine the amount of lipids and the concentration of EVs within one measurement. Importantly, NMR can also distinguish between EV‐derived and serum‐derived phospholipid signals, enabling contamination assessment in serum‐free systems (Mukerjee et al. 2025).

NMR metabolomics provides a dual analytical capability by simultaneously quantifying intrinsic EV lipid signals and extracellular metabolites reflecting cellular metabolic activity within the bioreactor. Of note, this enables the establishment of links between cell physiological state, EV production dynamics and functional bioactivity. For instance, variations in lipid classes (e.g. choline‐containing phospholipids) and key metabolic indicators (e.g. lactate, glucose and amino acid turnover) may reflect changes in EV biogenesis and cargo composition. Within a QbD framework, such metabolic and lipid signatures could be correlated with functional assays to define surrogate potency‐associated CQAs and support the development of predictive models of EV efficacy.

3.3. Enabling Quality by Design (QbD)

By integrating metabolomic data with CPPs (e.g. pH, oxygen and agitation), manufacturers can establish bioprocess design spaces and identify CQAs linked to EV performance. This alignment with Quality by Design (QbD) principles is essential for regulatory approval and technology transfer to industrial‐scale manufacturing. Beyond simple parameter mapping, QbD provides a structured framework that begins with defining a Quality Target Product Profile (QTPP) and systematically links measurable product attributes to safety and efficacy outcomes, as formalised in ICH Q8(R2) and increasingly adopted across advanced therapy manufacturing (Products CfHM 2017; Read et al. 2010).

Recent analyses of cell‐based therapeutic production demonstrate that biological complexity, variability in starting materials, and sensitivity to microenvironmental cues require a multivariate, risk‐based approach to process design, in which CQAs are identified early and iteratively refined through mechanistic understanding and empirical data (Lipsitz et al. 2016). Applying these principles to EV bioprocessing necessitates integrating high‐dimensional datasets (such as metabolomics, proteomics and particle‐level analytics) to uncover latent relationships between culture conditions, EV biogenesis pathways and functional potency.

Design‐of‐experiments (DoE) methodologies, combined with advanced small‐scale bioreactor platforms, enable efficient exploration of the multifactorial design space and robust identification of CPP interactions that would be missed by one‐factor‐at‐a‐time optimisation. This is consistent with findings from the cell therapy field, where DoE has been instrumental in revealing non‐intuitive interactions between seeding density, agitation, cytokine combinations and media exchange strategies, ultimately improving process robustness and cost‐effectiveness (Lipsitz et al. 2016). For EVs, similar approaches may help to elucidate how parameters such as shear stress, nutrient flux, hypoxia or bioreactor geometry influence vesicle cargo composition, subtype distribution and particle‐to‐protein ratios.

Furthermore, QbD supports the incorporation of PAT to enable real‐time monitoring and adaptive control of key variables. In cell therapy bioprocessing, PAT has already demonstrated value in controlling secreted signalling factors and improving product consistency. Translating this to EV manufacturing implies deploying online sensors for metabolites, near‐infrared or Raman spectroscopy for culture fingerprints, and real‐time particle analytics to ensure that the process remains within a validated design space (Rathore 2014). Such integrated control strategies enhance process understanding, reduce batch‐to‐batch variability and enable continuous improvement.

Overall, implementing a rigorous QbD strategy in EV bioprocessing not only aligns with regulatory expectations but also provides a scientific foundation for scalable, reproducible, and potency‐driven manufacturing of EV therapeutics.

4. The Galvano Approach: Integrating Metabolomics Into GMP‐Compliant EV Manufacturing Platform

The GALVANO (GMP‐Grade Analytical Laboratory Vesicles Application Novel Optimisation) consortium is developing Spain's first clinical‐grade (that is, produced under GMP conditions with defined release criteria) manufacturing platform for MSC‐EVs, termed EVCord. This initiative combines advanced bioprocess engineering with metabolomic analytics to enable reproducible and scalable EV production. GALVANO provides a uniquely integrated GMP‐grade manufacturing and PAT‐driven analytics platform, combining scalable bioreactor‐based MSC expansion with NMR‐powered metabolomic and lipidomic profiling, to deliver reproducible, clinically translatable EV products, thereby addressing one of the field's most pressing unmet needs: harmonised, evidence‐based criteria for large‐scale EV production and QC suitable for human therapeutic use.

4.1. Bioprocess Framework

The production of EVCord comprises three modular processes from clinical‐approved Wharton's jelly–derived MSCs (hMSC, WJ) cultured in 500 mL 3D single‐use STRs using chemically defined, serum‐free media (Lopez‐Fernandez et al. 2024) to a combination of TFF and SEC for downstream EV purification, ensuring high recovery and purity. Throughout the manufacturing procedure, we also implemented a QC framework aligned with GMP to identify key CPPs and CQAs in order to ensure EV batch‐to‐batch reproducibility. In brief, in‐process control measurements include: (i) cell viability, by conventional metabolic performance (i.e. glucose consumption and lactate formation); (ii) NMR profiling enabling the simultaneous monitoring of dozens of metabolites associated with cellular energetic function, including small molecules, ketone bodies and biomarkers of cellular health; (iii) particle size and concentration providing a reference for determining final product dosage; (iv) EV identity through a complete and comprehensive panel of surface EV‐specific markers due to their inherent relationship to EV biogenesis; (v) relative protein/lipid content to corroborate their composition and purity and to assess reproducibility between batches; (vi) potency, also to assess batch‐to‐batch reproducibility; and (vii) pH and safety, including endotoxin, sterility and adventitious viruses.

4.2. NMR Metabolomic Integration

GALVANO was designed to incorporate high‐throughput 1H‐NMR spectroscopy to: (i) monitor culture media composition throughout bioreactor runs; (ii) detect cellular metabolic shifts associated with stress or senescence; and (iii) characterise EV surface lipid signatures as markers of identity and stability. In this context, 1H‐NMR metabolomics enables real‐time monitoring of a broad panel of metabolites in culture media, including key intermediates of cellular energetic pathways, small‐molecule catabolites, ketone bodies, and markers related to cellular stress and viability. By capturing these molecular signatures directly from intact media without extraction, NMR provides a robust, low‐variability and high‐performance platform to characterise the metabolic state of hMSC and EV‐producing cultures throughout bioreactor operation, offering insight into nutrient consumption, energetic efficiency, redox balance and the onset of senescence or stress responses. In parallel, NMR also detects a membrane‐associated phospholipid signal characteristic of EVs, enabling the direct identification and quantification of EVs in conditioned media without additional purification steps. This reproducible and quantitative spectral feature supports batch‐to‐batch comparison and strengthens QC by tracking EV presence, yield and consistency across production runs.

Together, these complementary readouts integrate cellular metabolism with EV biogenesis monitoring, facilitating early deviation detection, automated decision‐making and optimisation of bioprocess performance.

4.3. From Pilot to Clinical Scale

The GALVANO roadmap envisions scaling production to 3 and 10 L bioreactors, enabling GMP manufacturing under European Medicines Agency (EMA) investigational biologics guidelines. The platform will support both preclinical testing in large, translational animal models and future clinical‐grade batch production for human trials. Indeed, recent advances in bioprocess engineering have demonstrated that MSC‐derived EVs can be reproducibly produced at scale using STRs under serum‐/xeno‐free, microcarrier‐based systems, yielding sufficient EV quantities for clinical demands while preserving characteristic EV identity markers, morphology and functional properties (Lopez‐Fernandez et al. 2024; Lopez‐Fernandez et al. 2024; Rafiq et al. 2016). However, critical challenges remain in defining and enforcing consistent release criteria (identity, purity, potency, sterility and batch‐to‐batch reproducibility) and in standardising upstream (cell expansion) and downstream (isolation and purification) processes amenable to GMP workflows. By combining scalable bioreactor‐based cell expansion with rigorous process control and standardised analytics, GALVANO stands to overcome many of these barriers, offering a harmonised, quality‐controlled manufacturing pipeline that can translate EV‐based therapies from bench to bedside in a regulatory‐compliant manner.

5. Challenges, Opportunities and Future Directions

The principal challenges affecting MSC‐derived EVs across development, GMP manufacturing and clinical application are summarised in Table 1, reflecting the complexity of translating EV bioprocesses into regulated settings. The advancement of metabolomics‐enabled bioprocessing for GMP‐grade EV manufacturing relies on deeply integrated, multidisciplinary collaboration. Designing robust workflows, validating them under GMP, and aligning them with evolving regulatory expectations requires coordinated expertise in cell biology, engineering, analytical chemistry, and clinical translation, expertise that no single institution can provide alone. Collaborative initiatives such as GALVANO illustrate how shared infrastructures, harmonised methodologies and collective problem‐solving accelerate innovation while safeguarding reproducibility, transparency and regulatory compliance. Within such frameworks, the development and optimisation of NMR‐based metabolomic protocols, advanced data‐analysis pipelines and rigorous SOPs play a central role in establishing reliable, quantitative monitoring tools for both cell culture performance and EV product consistency.

TABLE 1.

Cross‐cutting challenges on the development, manufacturing and use of MSC‐derived extracellular vesicles.

Bioprocess dimension Technical challenges Clinical‐scale challenges Quality control Ethics Regulatory compliance Ref.
Donor selection Donor diversity (age, health, collection conditions) affects MSC phenotype and EV cargo. Risks of adventitious agents. Sourcing large numbers of qualified donors while keeping traceability and screening capacity. Need for standardised donor metadata, documented screening (infectious agents, HLA, medications). Traceability requirements. Informed consent for donation and future EV products; donor privacy; equitable access to source tissues. Donor selection and testing must meet SoHO and ATMP requirements; documentation for GMP. Georgiev‐Hristov et al. (2022), Guell‐Alonso et al. (2025)
Cells Consistent cell identity, absence of transformation, control of passage/senescence. Culture adaption to serum‐free media. Producing large, consistent MCBs/WCBs with retained phenotype. Scale‐up impacts (senescence, altered secretome). Defined acceptance criteria (identity, potency surrogates, sterility, mycoplasma, adventitious agents, karyotype). Donor consent must permit long‐term banking and intended EV applications; ownership and benefit sharing. MCB creation under GMP with validated characterisation panel and documented comparability bridging studies for scaled banks. Tiwari et al. (2021), Oliver‐Vila et al. (2016), Grau‐Vorster et al. (2019)
Upstream (cell culture, EV release) Control of CPPs (DO, pH, shear stress, microcarrier); media composition (chemically defined vs serum) influences EV yield/composition. Bioreactor fouling, sampling. Scale‐dependent mixing, oxygen transfer and harvest logistics; reproducibility across batches and vessels (scale up/down). In‐process monitoring: metabolites, lactate/glucose, cell viability, real‐time bioreactor sensors; correlate metabolic fingerprints with EV yield (QbD). Concerns for use of serum for contamination risk; moving to xeno‐free media reduces ethical/regulatory burden. Process validation, closed single‐use systems preferred for GMP; CPPs documented; comparability studies for scale changes. Gandham et al. (2020), Staubach et al. (2021), Benevelli et al. (2022)
Downstream (clarification, concentration, purification) Method selection affects yield vs purity; vesicle integrity sensitive to shear and handling. Removal of protein aggregates and non‐vesicular particles is challenging. Scaling TFF/SEC while preserving recovery; column capacity, single‐use vs reusable components; process time vs stability. Orthogonal assays required: size (NTA, TRPS), morphology (cryo‐EM), surface markers (bead/flow), protein/RNA cargo, lipid profiling; assign acceptance criteria. Purity critical to avoid transferrable donor proteins/viruses; ethical duty to ensure safety before use. Downstream unit ops need GMP validation; viral clearance/sterility strategy; documentation for process changes. Gandham et al. (2020), Staubach et al. (2021), Vogel et al. (2021), Ni et al. (2024), Silva et al. (2021)
Analytics Lack of universal potency assay; surrogate assays (T‐cell inhibition, CD73 activity) may not capture all MoAs. Need for sensitive, reproducible assays. High throughput and standardised assays needed for many batches; sample throughput and automation concerns. Harmonised panels (MISEV recommendations) for size, markers, contaminants, potency; develop reference materials and interlaboratory ring trials. Transparency about assay limitations; patient safety paramount when potency uncertain. Regulators expect defined CQAs and validated potency assays; alignment with ISEV/EMA guidance. Ullah et al. (2021), Martin‐Lorenzo et al. (2023)
Formulation, storage, logistics EV stability is temperature and excipient dependent; freeze‐thaw affects integrity. Container interactions and adsorption issues. Scalable fill/finish under aseptic GMP; validated cold chain for transport to clinical sites; batch returns/recall plans. Release testing post‐thaw stability, particle recovery, potency maintenance; shelf‐life studies required. Equitable distribution, consent for storage/use; environmental considerations of single‐use plastics. Stability data and validated storage conditions required for regulatory dossiers; chain‐of‐custody documentation. Hengelbrock et al. (2025), Akbar et al. (2022)
Administration (route, dose, frequency) Route influences biodistribution (IV, intracoronary, local); formulation must be compatible with administration device; dosing units (particles, protein, activity). Clinical dosing at scale (multiple doses per patient) increases manufacturing demand; route‐specific safety profiles required. PK and PD readouts needed; establish surrogate biomarkers for exposure/response. Informed consent must reflect uncertainties in biodistribution and long‐term effects. Clinical trial design must justify dose metrics and safety monitoring; regulators expect dose‐finding and PK/PD data. Silva et al. (2021), de Jong et al. (2019), Skotland et al. (2022), Haghighitalab et al. (2025)
Patient follow‐up & safety monitoring Assay sensitivity for adverse event detection; long‐term surveillance for immunogenicity or off‐target effects. Large‐scale safety registries and infrastructure for multicentre follow‐up. Define safety endpoints, immunogenicity assays and standardised reporting templates. Transparency with patients on unknowns; data protection for long‐term follow‐up. Post‐market surveillance plans where applicable; centralised databases to capture outcomes/adverse events. Wardhani et al. (2024), Takakura et al. (2024)
Ethical, legal & socio‐economic aspects Patent/IP landscape can affect technical freedom to operate; data sharing vs IP tension. COGs and reimbursement strategies may limit clinical scalability and access. Need for open, FAIR datasets to allow reproducibility while protecting privacy/IP. Equity of access, benefit sharing with donors, and transparency in clinical claims. Regulators and payers will need health‐economic evidence; ethics boards review donor/recipient frameworks. Silva et al. (2021), Takakura et al. (2024), Stawarska et al. (2024)
Cross‐cutting: standardisation, reproducibility, collaboration Heterogeneous methods across labs hamper comparability (isolation, analytics). Multi‐site scale‐up requires harmonised SOPs and comparability protocols. Need ring trials, and MISEV‐aligned reporting to enable regulatory acceptance. Collaborative governance and transparency address ethical concerns and public trust. Harmonised regulatory engagement (early dialogue), cross‐sector consortia (academia‐industry‐regulators) and open method panels. Théry et al. (2018), Witwer et al. (2021), Börger et al. (2020), Konala et al. (2015)
Open in a new tab

Notes: This table synthesises the technical, regulatory and ethical hurdles associated with the EV medicinal product lifecycle. Literature Search Strategy: To ensure a transparent and broad evidence base for this Perspective, references were identified through a strategic search of PubMed and augmented by AI‐assisted discovery tools (ResearchRabbit and SciSpace). Selection prioritised studies at the intersection of NMR, MSC‐EV bioprocessing and GMP‐scale manufacturing with high translational relevance. Supporting empirical studies and bioprocessing reports are linked to each identified challenge in the References (Ref.) column to ensure traceability.

Abbreviations: ATMP, Advanced Therapy Medicinal Product; COG, cost of goods; CPP, critical process parameters; CQA, Critical Quality Attributes; cryo‐EM, cryo‐electron microscopy; DO, dissolved oxygen; EMA, European Medicines Agency; EV, extracellular vesicle; FAIR, findable, accessible, interoperable, and reusable; GMP, Good Manufacturing Practice; HLA, human leukocyte antigen; ISEV, International Society for Extracellular Vesicles; MCB, Master Cell Bank; MISEV, minimal information for studies of extracellular vesicles; MoA, mode of action; NTA, nanoparticle tracking analysis; PD, pharmacodynamics; PK, pharmacokinetics; QbD, Quality by Design; SEC, size‐exclusion chromatography; SoHO, substances of human origin; SOP, standard operating procedures; TFF, Tangential Flow Filtration; TRPS, tunable resistive pulse sensing; WCB, Working Cell Bank.

To achieve meaningful clinical translation, the EV field must urgently converge on standardised reporting frameworks that integrate metabolic and bioprocess variables with conventional EV characterisation. These frameworks should define core elements such as culture conditions, media composition, dynamic versus static expansion systems and feeding strategies; appropriate metabolic monitoring techniques and reference metabolite panels; harmonised protocols for NMR sample preparation, acquisition, and data processing; and shared, well‐annotated repositories for EV metabolomic datasets. Establishing consensus around these elements is essential to improve cross‐laboratory comparability, reduce variability and strengthen confidence in EV‐derived medicinal products.

Regulatory authorities are increasingly emphasising the need for integrated analytical strategies that demonstrate process robustness, batch‐to‐batch consistency, and the mechanistic basis of EV identity, purity and potency. Harmonisation between leading scientific societies (namely ISEV and ISCT) and EMA frameworks, supported by collaborative pilot studies, will be critical for defining minimum analytical standards suitable for investigational and, ultimately, commercial use. Quantitative and highly reproducible metabolomics, particularly when enabled by NMR‐based platforms, is uniquely positioned to serve as a bridging technology across academic, industrial, and regulatory domains, facilitating the transition from exploratory research to regulated manufacturing environments.

Looking ahead, the growth of high‐resolution metabolomic datasets will enable the application of machine‐learning models capable of predicting EV yield, composition, and therapeutic fitness based on real‐time metabolic signals. Such models have the potential to deliver adaptive bioprocess control, automated anomaly detection, and predictive maintenance in GMP‐compliant environments. Progress will also depend on the development of standard reference materials (e.g. lyophilised EV preparations with certified metabolite or phospholipid profiles) to support inter‐laboratory benchmarking and long‐term quality surveillance. Furthermore, integrating metabolomics with proteomics and transcriptomics will provide a mechanistic, multi‐layered understanding of how cellular physiology shapes EV cargo and function, supporting rational optimisation of manufacturing processes and potency‐defining signatures. Ultimately, sustained collaboration between academic groups, clinical researchers, industry partners, and regulatory bodies will be essential to ensure that emerging EV‐based therapies are scalable, affordable, and aligned with regulatory expectations. The GALVANO model, grounded in early coordination across these sectors, provides a robust template for accelerating translation and enabling future clinical adoption.

6. Final Remarks

The transition of EV‐based therapies from promising preclinical tools to approved clinical products hinges on the development of standardised, scalable and analytically transparent bioprocesses. Metabolomics and more specifically NMR‐based profiling offers a novel and practical means to achieve real‐time process monitoring, predictive quality assessment and enhanced batch consistency. Through initiatives such as the GALVANO consortium, EV researchers propose a roadmap for integrating powerful analytical tools such as to advance in metabolomics analyses into the full lifecycle of EV manufacturing, from upstream cell culture to final product validation. We further advocate for collaborative networks, shared data infrastructures and harmonised analytical standards to drive the field toward regulatory acceptance and clinical success. Importantly, recent critical evaluations of current MISEV‐recommended lipid quantification methods have highlighted methodological limitations in accurately capturing total lipid content in EV preparations. In this context, we propose that NMR‐based metabolomics may represent a valuable complementary analytical approach to existing techniques, contributing to more robust, quantitative and reproducible characterisation frameworks. Future updates of EV characterisation guidelines may benefit from incorporating such orthogonal technologies to better reflect the complexity of EV‐associated lipid and metabolic signatures.

In our view, the next decade of EV manufacturing will be defined not solely by technological advances but also by the collective ability of the community to collaborate, standardise and harmonise. By embedding metabolomics into the design and regulation of EV production, we can accelerate the arrival of more safe, effective and affordable EV‐based therapeutics.

Author Contributions

Santiago Roura: conceptualization, investigation, funding acquisition, methodology, writing – original draft, writing – review and editing. Núria Amigó: conceptualization, investigation, funding acquisition, writing – original draft, methodology, writing – review and editing. Joaquim Vives: conceptualization, investigation, writing – original draft, funding acquisition, methodology, writing – review and editing, project administration.

Funding

This research was especially funded by the Spanish Ministerio de Ciencia Innovación y Universidades ref. CPP2023‐010430. BST is a member of Red Española de Terapias Avanzadas (RICORS‐TERAV/TERAV+, expedient no.’s RD21/0017/0022 and RD24/0014/0037) funded by the Instituto de Salud Carlos III (ISCiii) in the context of NextGenerationEU's Recovery, Transformation and Resilience Plan. This publication is based upon work from COST Action No. CA24114 (BTCs4ATMP: European network of SoHO establishments for rapid and sustainable access to ATMPs), supported by COST (European Cooperation in Science and Technology), www.cost.eu/actions/CA24114. The joint BST‐VHIR Musculoskeletal Tissue Engineering group is a Consolidated Research Group by the Agència de Gestió d'Ajuts Universitaris i de Recerca (AGAUR) of the Generalitat de Catalunya (expedient no.’s 2021‐SGR‐00877). The ICREC Research Program was supported by grants from CIBERCV (CB16/11/00403) as part of the Plan Nacional de I+D+I, Generalitat de Catalunya (2021 SGR‐01437), and the JMC Legacy Research Fund of Germans Trias i Pujol University Hospital (2021_44).

Consent

Not applicable.

Conflicts of Interest

The authors declare no potential conflicts of interest.

Institutional Review Board Statement

Not applicable.

Declaration of Generative AI and AI‐Assisted Technologies in the Writing Process

During the preparation of this work, the authors used ChatGPT in order to improve the clarity and linguistic accuracy of the manuscript. Then, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

Acknowledgements

The authors gratefully acknowledge all members of our laboratories who contributed to the conception and strategic development of the GALVANO project described here, a project jointly coordinated by three research groups, whose insights and collaborative efforts were instrumental to its design.

Contributor Information

Santiago Roura, Email: sroura@igtp.cat.

Núria Amigó, Email: namigo@biosferteslab.com.

Joaquim Vives, Email: jvives@bst.cat.

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.

References

  1. Akbar, A. , Malekian F., Baghban N., Kodam S. P., and Ullah M.. 2022. “Methodologies to Isolate and Purify Clinical Grade Extracellular Vesicles for Medical Applications.” Cells 11, no. 2: 186. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alexandre, L. , Sun J., Taverna M., and Zhong W.. 2023. “Advances in Extracellular Vesicle Analysis.” Analytical and Bioanalytical Chemistry 415, no. 7: 1235–1238. [DOI] [PubMed] [Google Scholar]
  3. Bauer, F. N. , Tertel T., Stambouli O., et al. 2023. “CD73 Activity of Mesenchymal Stromal Cell‐Derived Extracellular Vesicle Preparations Is Detergent‐Resistant and Does Not Correlate With Immunomodulatory Capabilities.” Cytotherapy 25, no. 2: 138–147. [DOI] [PubMed] [Google Scholar]
  4. Benevelli, F. , Vella S., Crosta C., et al. 2022. “NMR as Powerful Technology for Noninvasively Monitoring Cell Health and Expansion During Bioprocessing.” Biotechnology and Bioengineering 119, no. 12: 3497–3508. [DOI] [PubMed] [Google Scholar]
  5. Böing, A. N. , van der Pol E., Grootemaat A. E., Coumans F. A., Sturk A., and Nieuwland R.. 2014. “Single‐Step Isolation of Extracellular Vesicles by Size‐Exclusion Chromatography.” Journal of Extracellular Vesicles 3, no. 1: 23430. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Börger, V. , Weiss D. J., Anderson J. D., et al. 2020. “International Society for Extracellular Vesicles and International Society for Cell and Gene Therapy Statement on Extracellular Vesicles From Mesenchymal Stromal Cells and Other Cells: Considerations for Potential Therapeutic Agents to Suppress Coronavirus Disease‐19.” Cytotherapy 22, no. 9: 482–485. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Busatto, S. , Vilanilam G., Ticer T., et al. 2018. “Tangential Flow Filtration for Highly Efficient Concentration of Extracellular Vesicles From Large Volumes of Fluid.” Cells 7, no. 12: 273. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Chen, J. , Li P., Zhang T., et al. 2021. “Review on Strategies and Technologies for Exosome Isolation and Purification.” Frontiers in Bioengineering and Biotechnology 9: 811971. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. CHMP . 2017. “ICH Guideline Q8 (R2) on Pharmaceutical Development Step 5.” In Products Committee for Human Medicinal Products, edited by European Medicines Agency, 24. European Union. [Google Scholar]
  10. Clos‐Sansalvador, M. , Monguió‐Tortajada M., Roura S., Franquesa M., and Borràs F. E. 2022. “Commonly Used Methods for Extracellular Vesicles' Enrichment: Implications in Downstream Analyses and Use.” European Journal of Cell Biology 101, no. 3: 151227. [DOI] [PubMed] [Google Scholar]
  11. Coumans, F. A. W. , Brisson A. R., Buzas E. I., et al. 2017. “Methodological Guidelines to Study Extracellular Vesicles.” Circulation Research 120, no. 10: 1632–1648. [DOI] [PubMed] [Google Scholar]
  12. de Jong, O. G. , Kooijmans S. A. A., Murphy D. E., et al. 2019. “Drug Delivery With Extracellular Vesicles: From Imagination to Innovation.” Accounts of Chemical Research 52, no. 7: 1761–1770. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gámez‐Valero, A. , Monguió‐Tortajada M., Carreras‐Planella L., Franquesa M., Beyer K., and Borràs F. E. 2016. “Size‐Exclusion Chromatography‐based Isolation Minimally Alters Extracellular Vesicles' Characteristics Compared to Precipitating Agents.” Scientific Reports 6: 33641. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Gandham, S. , Su X., Wood J., et al. 2020. “Technologies and Standardization in Research on Extracellular Vesicles.” Trends in Biotechnology 38, no. 10: 1066–1098. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Georgiev‐Hristov, T. , Garcia‐Arranz M., Trebol‐Lopez J., Barba‐Recreo P., and Garcia‐Olmo D.. 2022. “Searching for the Optimal Donor for Allogenic Adipose‐Derived Stem Cells: A Comprehensive Review.” Pharmaceutics 14, no. 11: 2338. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Grau‐Vorster, M. , Rodriguez L., Del Mazo‐Barbara A., et al. 2019. “Compliance With Good Manufacturing Practice in the Assessment of Immunomodulation Potential of Clinical Grade Multipotent Mesenchymal Stromal Cells Derived From Wharton's Jelly.” Cells 8, no. 5: 484. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Guell‐Alonso, R. , Cabrera‐Perez R., and Vives J.. 2025. “Immunocompatibility in Transplantation: Adapting to a Changing Therapeutic Landscape.” Frontiers in Immunology 16: 1648534. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Haghighitalab, A. , Kazemi Noughabi M., and Hassanzadeh H. et al. 2025. “Developing a General Checklist for the Effective Administration of Extracellular Vesicles in Biomedical and Clinical Research.” Cell Journal 26, no. 11: 656–668. [DOI] [PubMed] [Google Scholar]
  19. Hengelbrock, A. , Baukmann S., Uhl A., Schmidt A., and Strube J.. 2025. “Digital Twin for the Production of hMSC‐Derived Extracellular Vesicles for Applications in Cell and Gene Therapy Toward Autonomous Operation.” ACS Omega 10, no. 40: 46659–46681. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Jiang, X. , Yang J., Lin Y., et al. 2023. “Extracellular Vesicles Derived From human ESC‐MSCs Target Macrophage and Promote Anti‐Inflammation Process, Angiogenesis, and Functional Recovery in ACS‐induced Severe Skeletal Muscle Injury.” Stem Cell Research & Therapy 14, no. 1: 331. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Konala, V. B. , Mamidi M. K., Bhonde R., Das A. K., Pochampally R., and Pal R.. 2015. “The Current Landscape of Mesenchymal Stromal Cell Secretome: A New Paradigm for Cell‐Free Regeneration.” Cytotherapy 18, no. 1: 13–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lener, T. , Gimona M., Aigner L., et al. 2015. “Applying Extracellular Vesicles Based Therapeutics in Clinical Trials—An ISEV Position Paper.” Journal of Extracellular Vesicles 4: 30087. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Li, M. , Fang F., Sun M., Zhang Y., Hu M., and Zhang J.. 2022. “Extracellular Vesicles as Bioactive Nanotherapeutics: An Emerging Paradigm for Regenerative Medicine.” Theranostics 12, no. 11: 4879–4903. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lipsitz, Y. Y. , Timmins N. E., and Zandstra P. W. 2016. “Quality Cell Therapy Manufacturing by Design.” Nature Biotechnology 34, no. 4: 393–400. [DOI] [PubMed] [Google Scholar]
  25. Lopez‐Fernandez, A. , Codinach M., Coca M. I., et al. 2024. “Comparability Exercise of Critical Quality Attributes of Clinical‐Grade human Mesenchymal Stromal Cells From the Wharton's Jelly: Single‐Use Stirred Tank Bioreactors versus Planar Culture Systems.” Cytotherapy 26, no. 5: 418–426. [DOI] [PubMed] [Google Scholar]
  26. Lopez‐Fernandez, A. , Garcia‐Gragera V., Lecina M., and Vives J.. 2024. “Identification of Critical Process Parameters for Expansion of Clinical Grade human Wharton's Jelly‐Derived Mesenchymal Stromal Cells in Stirred‐Tank Bioreactors.” Biotechnology Journal 19, no. 2: e2300381. [DOI] [PubMed] [Google Scholar]
  27. Martin‐Lorenzo, M. , Molero D., and Alvarez‐Llamas G.. 2023. “Metabolomics Analysis of Urinary Extracellular Vesicles by Nuclear Magnetic Resonance and Liquid Chromatography‐Mass Spectrometry.” Methods in Molecular Biology 2668: 57–68. [DOI] [PubMed] [Google Scholar]
  28. Mukerjee, N. , Bhattacharya A., Maitra S., et al. 2025. “Exosome Isolation and Characterization for Advanced Diagnostic and Therapeutic Applications.” Materials Today Bio 31: 101613. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Munoz‐Dominguez, N. , Roura S., Prat‐Vidal C., and Vives J.. 2021. “Wharton's Jelly Mesenchymal Stromal Cells and Derived Extracellular Vesicles as Post‐Myocardial Infarction Therapeutic Toolkit: An Experienced View.” Pharmaceutics 13, no. 9: 1336. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Nagana Gowda, G. A. , Zhu W., and Raftery D.. 2025. “NMR‐based Metabolomics: Where Are We Now and Where Are We Going?” Progress in Nuclear Magnetic Resonance Spectroscopy 150–151: 101564. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Ni, F. , Zhu Q., Li H., Liu F., and Chen H.. 2024. “Efficient Preparation of High‐Purity and Intact Mesenchymal Stem Cell‐Derived Extracellular Vesicles.” Analytical and Bioanalytical Chemistry 416, no. 8: 1797–1808. [DOI] [PubMed] [Google Scholar]
  32. Oliver‐Vila, I. , Coca M. I., Grau‐Vorster M., et al. 2016. “Evaluation of a Cell‐Banking Strategy for the Production of Clinical Grade Mesenchymal Stromal Cells From Wharton's Jelly.” Cytotherapy 18, no. 1: 25–35. [DOI] [PubMed] [Google Scholar]
  33. Rafiq, Q. A. , Coopman K., Nienow A. W., and Hewitt C. J. 2016. “Systematic Microcarrier Screening and Agitated Culture Conditions Improves Human Mesenchymal Stem Cell Yield in Bioreactors.” Biotechnology Journal 11, no. 4: 473–486. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Rathore, A. S. 2014. “QbD/PAT for Bioprocessing: Moving From Theory to Implementation.” Current Opinion in Chemical Engineering 6: 1–8. [Google Scholar]
  35. Read, E. K. , Shah R. B., Riley B. S., Park J. T., Brorson K. A., and Rathore A. S. 2010. “Process Analytical Technology (PAT) for Biopharmaceutical Products: Part II. Concepts and Applications.” Biotechnology and Bioengineering 105, no. 2: 285–295. [DOI] [PubMed] [Google Scholar]
  36. Roura, S. , Galvez‐Monton C., Mirabel C., Vives J., and Bayes‐Genis A.. 2017. “Mesenchymal Stem Cells for Cardiac Repair: Are the Actors Ready for the Clinical Scenario?” Stem Cell Research & Therapy 8, no. 1: 238. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Sheng, Q. , Wu Z., Li W., et al. 2025. “Extracellular Vesicles in Metabolic Perspective: Mechanism and Targeted Therapy.” Journal of Nanobiotechnology 23, no. 1: 676. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Silva, A. K. A. , Morille M., Piffoux M., et al. 2021. “Development of Extracellular Vesicle‐based Medicinal Products: A Position Paper of the Group “Extracellular Vesicle translatiOn to clinicaL perspectiVEs—EVOLVE France”.” Advanced Drug Delivery Reviews 179: 114001. [DOI] [PubMed] [Google Scholar]
  39. Skotland, T. , Ekroos K., Llorente A., and Sandvig K.. 2025. “Quantitative Lipid Analysis of Extracellular Vesicle Preparations: A Perspective.” Journal of Extracellular Vesicles 14, no. 3: e70049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Skotland, T. , Iversen T. G., Llorente A., and Sandvig K.. 2022. “Biodistribution, Pharmacokinetics and Excretion Studies of Intravenously Injected Nanoparticles and Extracellular Vesicles: Possibilities and Challenges.” Advanced Drug Delivery Reviews 186: 114326. [DOI] [PubMed] [Google Scholar]
  41. Staubach, S. , Bauer F. N., Tertel T., et al. 2021. “Scaled Preparation of Extracellular Vesicles From Conditioned media.” Advanced Drug Delivery Reviews 177: 113940. [DOI] [PubMed] [Google Scholar]
  42. Stawarska, A. , Bamburowicz‐Klimkowska M., Runden‐Pran E., et al. 2024. “Extracellular Vesicles as Next‐Generation Diagnostics and Advanced Therapy Medicinal Products.” International Journal of Molecular Sciences 25, no. 12: 6533. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Takakura, Y. , Hanayama R., Akiyoshi K., et al. 2024. “Quality and Safety Considerations for Therapeutic Products Based on Extracellular Vesicles.” Pharmaceutical Research 41, no. 8: 1573–1594. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Ter‐Ovanesyan, D. , Norman M., Lazarovits R., et al. 2021. “Framework for Rapid Comparison of Extracellular Vesicle Isolation Methods.” Elife 10: e70725. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Théry, C. , Witwer K. W., Aikawa E., et al. 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, no. 1: 1535750. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Tiwari, S. , Kumar V., Randhawa S., and Verma S. K. 2021. “Preparation and Characterization of Extracellular Vesicles.” American Journal of Reproductive Immunology 85, no. 2: e13367. [DOI] [PubMed] [Google Scholar]
  47. Ullah, M. S. , Zhivonitko V. V., Samoylenko A., et al. 2021. “Identification of Extracellular Nanoparticle Subsets by Nuclear Magnetic Resonance.” Chemical Science 12, no. 24: 8311–8319. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Van Deun, J. , Mestdagh P., Agostinis P., et al. 2017. “EV‐TRACK: Transparent Reporting and Centralizing Knowledge in Extracellular Vesicle Research.” Nature Methods 14, no. 3: 228–232. [DOI] [PubMed] [Google Scholar]
  49. Vogel, R. , Savage J., Muzard J., et al. 2021. “Measuring Particle Concentration of Multimodal Synthetic Reference Materials and Extracellular Vesicles With Orthogonal Techniques: Who Is up to the Challenge?” Journal of Extracellular Vesicles 10, no. 3: e12052. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Wang, J. , Zhang M., and Wang H.. 2024. “Emerging Landscape of Mesenchymal Stem Cell Senescence Mechanisms and Implications on Therapeutic Strategies.” ACS Pharmacology & Translational Science 7, no. 8: 2306–2325. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Wardhani, K. , Levina A., Grau G. E. R., and Lay P. A. 2024. “Fluorescent, Phosphorescent, Magnetic Resonance Contrast and Radioactive Tracer Labelling of Extracellular Vesicles.” Chemical Society Reviews 53, no. 13: 6779–6829. [DOI] [PubMed] [Google Scholar]
  52. Witwer, K. W. , Goberdhan D. C., and O'Driscoll L., et al. 2021. “Updating MISEV: Evolving the Minimal Requirements for Studies of Extracellular Vesicles.” Journal of Extracellular Vesicles 10, no. 14: e12182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Wu, Q. , Wang J., Tan W. L. W., et al. 2020. “Extracellular Vesicles From Human Embryonic Stem Cell‐Derived Cardiovascular Progenitor Cells Promote Cardiac Infarct Healing Through Reducing Cardiomyocyte Death and Promoting Angiogenesis.” Cell Death & Disease 11, no. 5: 354. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Yáñez‐Mó, M. , Siljander P. R., Andreu Z., et al. 2015. “Biological Properties of Extracellular Vesicles and Their Physiological Functions.” Journal of Extracellular Vesicles 4: 27066. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Zhang, J. , Li J., Zheng Y., et al. 2025. “Lipid Metabolism and Lipid Signaling in Extracellular Vesicles Ontogeny: From Biogenesis to Functional Execution.” Journal of Nanobiotechnology 23, no. 1: 672. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Zhu, L. , Wang Q., Guo M., et al. 2024. “Mesenchymal Stem Cell‐Derived Exosomes in Various Chronic Liver Diseases: Hype or Hope?” Journal of Inflammation Research 17: 171–189. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.

Articles from Journal of Extracellular Vesicles are provided here courtesy of Wiley

RESOURCES