Extracellular vesicles (EVs) are small, membrane-enclosed particles secreted by cells into the surrounding extracellular environment, mediating the transport of functional molecules such as proteins, lipids, and nucleic acids [1]. They play a crucial role in intercellular communication and in modulating both physiological and pathological processes [2]. EVs are classified as exosomes, microvesicles, or apoptotic bodies according to their biological origin [3]. Inconsistent use of EV-related terminology can lead to confusion and misinterpretation in the literature [4]. To address this, the Minimal Information for Studies of Extracellular Vesicles 2023 guidelines recommend using the general term “EVs” unless the specific biogenesis pathway is identified [5]. For example, exosomes originate from multivesicular bodies (MVBs), while microvesicles derive from the plasma membrane. When classification is based on size and isolation methods, terms such as small EVs (less than 200 nm) and large EVs (greater than 200 nm) are appropriate.

The process of EV biogenesis begins with endocytosis, in which the cell membrane folds inward to form endosomes, also known as intraluminal vesicles, that encapsulate proteins, lipids, and nucleic acids [6]. Through successive inward invaginations, the membranes of early endosomes mature into late endosomes [6]. MVBs, a type of late endosome, contain multiple internal vesicles, each enclosing a portion of the cytosol along with various nucleic acids and proteins [7]. MVBs can then fuse with lysosomes, leading to degradation, or with the plasma membrane, resulting in the release of EVs into the extracellular space (Fig. 1) [8,9].

Mesenchymal stem cells (MSCs), derived from adipose tissue, umbilical cord, bone marrow, and dental pulp, are widely recognized for their therapeutic potential [10–12]. However, although 2-dimensional (2D) monolayer culture remains the gold standard for expansion, it does not adequately replicate the native in vivo microenvironment. 2D stem cell culture negatively affects cell growth kinetics, multipotency, chromosomal stability, and differentiation, and accelerates cellular senescence [13–15]. By contrast, 3-dimensional (3D) cell culture platforms, including scaffold-free (spheroid formation) and scaffold-based approaches, more closely mimic in vivo conditions [16]. With ongoing innovations in biomaterials, various 3D-based structures such as hydrogels, nanofibers, and cryogels are being investigated for MSC culture. Compared to 2D cultures, extracellular matrix (ECM)-based hydrogels markedly enhance the production of cellular cytokines and growth factors, thereby improving cellular functionality [17]. The oriented architecture of nanofibers further promotes the synthesis of pro-angiogenic and anti-inflammatory factors [18]. Similar to in vivo conditions, MSCs cultured in 3D systems exhibit greater production of osteogenic and angiogenic factors [19]. 3D cell culture systems have been utilized to explore EVs as potential biomarkers, revealing distinct molecular signaling and protein profiles [16,20,21]. Additionally, 3D-cultured EVs demonstrate enhanced therapeutic properties, including increased anti-inflammatory and pro-angiogenic activities [22–24]. Furthermore, 3D culture methods enable large-scale EV production, which is essential for the clinical translation of EV-based therapies [25].

Organoids, an advanced 3D culture platform, can be derived from induced pluripotent stem cells (iPSCs) or from adult stem cells (ASCs) isolated from tissue. Tissue-derived organoids, cultured from healthy or diseased donor stem cells, retain the characteristics of the source tissue, making them valuable for studying tissue growth mechanisms and for disease modeling [26,27]. As a physiologically relevant and customizable system, organoids bridge the gap between 2D cell cultures and animal models, providing more accurate tissue replication through enhanced cell–cell interactions [28]. By overcoming the limitations of 2D cultures, organoid models offer a powerful platform for drug discovery and disease research [29].

Most studies have focused on EV isolation from 2D culture dishes (2D-EVs) or from scaffold-free or scaffold-based 3D cultures (3D-EVs). As 3D cell clusters, organoids support more complex intercellular communication, enhancing their biological relevance. Consequently, organoid-derived EVs (OEVs) display greater complexity than 2D-EVs. Retaining the nanoscale size and biocompatibility of EVs while being richly enriched in bioactive molecules, OEVs hold significant promise for medical applications.

This review presents a comparison between 2D-EVs and 3D-EVs, emphasizing their respective advantages, limitations, and therapeutic applications. In addition, we examine recent advances in OEVs, investigating their unique properties compared to both 2D-EVs and 3D-EVs. Through a critical assessment of OEV strengths and challenges, we aim to provide valuable insights into their potential roles in precision medicine, biomarker discovery, and next-generation therapeutics.

Ethics statement: This study was a literature review of previously published studies and was therefore exempt from institutional review board approval.

A reliable and abundant supply of MSCs, the primary source of EVs, is essential to maximize their therapeutic potential. However, the low yield of 2D-EVs and limitations inherent to conventional culture systems pose significant challenges. Traditional methods for harvesting therapeutic EVs rely on 2D culture systems, in which MSCs grow as a monolayer on a substrate. Conventional 2D culture of MSCs accelerates cellular senescence, diminishing proliferation, differentiation, migration, and angiogenesis [30–32]. In 2D culture, cells lack polarity, which alters responses to processes such as apoptosis, while impaired external interactions further disrupt internal structural organization [33–35], as shown in Fig. 2 [35]. Moreover, 2D cultures do not replicate the native MSC niche, significantly limiting the therapeutic potential of EVs for clinical applications. To address these hurdles, innovative preconditioning strategies—including 3D culture and hypoxia—have been developed to enhance both EV production and functionality [24]. To preserve MSC stemness and improve EV efficacy, 3D culture offers a more physiologically relevant alternative [36]. The 3D culture platform increases both the yield and the functional quality of EV cargo, ultimately improving MSC functionality [37].

3D-EVs are enriched with proteins and miRNAs that play critical roles in energy metabolism, immune regulation, bone regeneration, cell proliferation, spinal cord development, and neuronal function [23,38]. Compared to 2D-EVs, they contain higher concentrations of anti-inflammatory and anti-apoptotic factors, indicating superior therapeutic potential [39]. A hollow fiber bioreactor-based 3D culture system enables high-yield production of EVs purified from umbilical cord-derived MSCs (UC-MSCs) [37]. These 3D-EVs display enhanced cardioprotective effects compared to those from 2D cultures. Similarly, another study demonstrated that MSC-derived EVs produced using a hollow fiber bioreactor-based 3D system achieved high yield and improved therapeutic efficacy, especially in protecting kidney function by mitigating cisplatin-induced acute kidney injury [24].

Previous reports have demonstrated that 3D cell culture systems—both scaffold-free and scaffold-based—more closely replicate in vivo conditions, thereby enhancing EV secretion and functionality [40]. Various 3D culture platforms, including hydrogels, cryogels, and nanofibers, have been investigated, integrating advances in biomaterials. Additionally, ECM-containing hydrogels promote increased secretion of growth factors and cytokines, providing stronger support for cellular function compared to 2D cultures [17]. D-EVs derived from UC-MSCs cultured in a hollow fiber 3D system show significantly higher yields than 2D-EVs [37]. Additionally, 3D-EVs display superior cardioprotective effects in acute myocardial infarction rats and are more effective in reducing cardiomyocyte apoptosis and enhancing angiogenesis compared to 2D-EVs. In another report, 3D-EVs from human UC-MSCs cultured on a 3D graphene scaffold outperformed 2D-EVs in reducing Aβ accumulation and improving spatial learning and memory in APP/PS1 transgenic mice [41]. This enhanced therapeutic effect correlates with higher levels of Aβ-degrading enzymes (NEP, IDE, and HSP70), which promote Aβ clearance and reduce inflammation and oxidative stress.

3D-EVs exhibit distinct molecular signaling and protein profiles compared to 2D-EVs [16,20]. Multiple studies have demonstrated the enhanced therapeutic efficacy of 3D culture-derived EVs, including increased pro-angiogenic, regenerative, and anti-inflammatory activities [22–24]. Furthermore, 3D culture techniques enable efficient cell expansion, optimizing space and cost, which leads to higher EV yields—a critical factor for the clinical application of EVs across various disease therapies [25].

A study assessed the impact of 3D dynamic wave-motion bioreactor culture on human MSC-derived EV production and properties compared to 2D culture [42]. The 3D system increased EV yield by approximately twofold and produced smaller EVs than the 2D platform. Notably, 3D-EVs exhibited upregulation of EV biogenesis-related genes, including both ESCRT-dependent (Alix, TSG, Rab27B) and ESCRT-independent (CD81, CD9, CD63) pathways. Another study reported a 24-fold increase in EV yield from human bone marrow-derived MSCs cultured in 3D vertical-wheel bioreactors with microcarriers compared to 2D cultures [43]. Additionally, these 3D-EVs demonstrated superior efficacy in promoting neurite outgrowth.

Culturing adipose-derived MSCs in a 3D bioreactor system (VITVO; Rigenerand) significantly increased both EV yield and purity compared to 2D culture [44]. One report found higher levels of HGF and ICAM-1 in EVs derived from MSCs cultured on 3D electrospun fiber scaffolds [45]. These 3D-EVs also markedly enhanced corneal wound healing in vitro and ex vivo, compared to 2D-EVs. Furthermore, 3D-EVs derived from BM-MSCs cultured on a 3D collagen scaffold significantly improved spatial learning in traumatic brain injury rats versus 2D-EVs [46]. This superior therapeutic effect is attributed to their ability to reduce inflammation and promote both angiogenesis and neurogenesis.

Collectively, extensive research demonstrates the superiority of 3D culture platforms in producing higher yields of EVs with significantly greater therapeutic potential compared to 2D culture techniques. 3D cell cultures enhance cell–cell interactions and intercellular communication relative to 2D cultures, contributing to the superior properties of 3D-EVs. However, for clinical application of 3D-EVs, it remains crucial to optimize MSC culture conditions to maximize EV yield and improve therapeutic efficacy [47,48]. Unlike standard 3D cultures, organoids promote greater cell diversity and more complex intercellular communication, suggesting that OEVs may possess even greater functional complexity than 3D-EVs. The following sections discuss the production of EVs within these advanced 3D cellular structures—organoids.

Organoids are 3D cell structures that self-assemble and differentiate within an in vitro 3D environment, closely recapitulating the spatial organization and functional properties of native organ-like structures [49–51]. ASCs, iPSCs, and embryonic stem cells (ESCs) all possess the capacity to generate organoids [52,53]. However, cell sorting and spatially restricted lineage commitment can limit self-assembly during organoid development [54]. Successful organoid formation depends on specific cytokines and optimal culture conditions, which serve as essential regulatory factors [55].

Adult tissues, such as skin and blood, are considered the primary source of ASCs [56,57]. Due to their limited differentiation potential, ASCs can generate organoids with functionality restricted to cell types present in the donor tissue [58,59]. Nevertheless, ASCs are advantageous due to their accessibility and widespread distribution, making them practical for organoid research. In contrast, iPSCs and ESCs exhibit broader differentiation capacity, enabling development into a wide variety of cell types and the simulation of diverse organ structures [60,61]. This versatility allows iPSC-derived organoids to achieve greater structural complexity and functional diversity, replicating a broader spectrum of physiological functions and more intricate spatial organization [62,63]. For those reasons, the PSC-derived organoids culture platform is considered reliable for organoid-associated disease modeling and novel drug discovery. For these reasons, pluripotent stem cell (PSC)-derived organoid culture platforms are considered reliable tools for organoid-based disease modeling and innovative drug discovery [64,65].

Despite recent advances, organoid research remains in its early stages, and most current organoids exhibit limited functionality, falling short of the goal of fully multi-functional tissue models. The overarching aim of organoid technology is to develop multi-functional organoids, thereby improving therapeutic outcomes [66]. One potential application of organoids is the isolation of EVs or OEVs.

Both OEVs and conventional EVs are classified as mammalian EVs and share similar origins, biogenesis pathways, molecular components, and internalization mechanisms. However, OEVs and EVs differ due to the organoid 3D formation process and the cellular complexity introduced by the presence of multiple cell types within organoids. As a result, OEVs more closely resemble human body fluid-derived EVs [54]. Incorporating ECM-containing scaffolds, such as Matrigel, in organoid generation presents a significant challenge for OEV isolation, making the process more complex than the isolation of conventional EVs. Therefore, OEV isolation and purification protocols must be carefully designed to prevent contamination and maintain the integrity and functional properties of the isolated OEVs.

Several methods are available for OEV isolation, each with unique advantages and drawbacks. Ultracentrifugation is widely used due to its high yield and practicality, but its limitations include low purity and prolonged processing time [67]. Differential centrifugation is cost-effective and straightforward but results in low-purity OEVs and requires lengthy processing [68]. Density gradient centrifugation offers high purity and yield, but it is time-consuming, costly, and requires specialized equipment [69]. Ultrafiltration is an efficient and high-yield method, though it can be hampered by membrane blockage [70]. The immunoaffinity-based capture method achieves high purity and specificity, but it is expensive and hampered by the lack of well-defined markers for OEV identification [71]. Size exclusion chromatography is easy to use and provides high purity, yet it is costly and time-intensive [72].

To avoid external EV contamination in organoid cultures, it is critical to use EV-free media supplemented with EV-depleted serum [73–75].

The structural complexity of organoids and the presence of multiple cell types make OEV characterization more challenging than for conventional EVs. While transmission electron microscopy and nanoparticle tracking analysis are commonly employed to assess morphology, size, and concentration, the heterogeneity of OEVs necessitates more meticulous analysis. Western blotting and fluorescent surface marker detection must be tailored to the cellular origin of OEVs to ensure accurate characterization [76]. Ultimately, selecting the optimal OEV isolation method requires balancing purity, yield, cost, and processing time to align with research or clinical objectives.

Recent preclinical and clinical studies have demonstrated the therapeutic potential, drug delivery capability, and diagnostic value of EVs. However, low EV yield remains a major challenge for clinical translation, prompting extensive research into strategies for enhancing EV production [9]. 3D stem cell culture offers significant advantages over 2D culture, more accurately replicating the native microenvironment of cells. 3D culture systems enhance cell–cell and cell–matrix interactions, which are crucial for maintaining cellular integrity, differentiation potential, and functional activity [24,84]. Additionally, the 3D microenvironment facilitates activation of key cellular signaling pathways, promoting stemness, survival, and regenerative capacity [40]. Notably, 3D-cultured stem cells secrete higher levels of anti-inflammatory and pro-angiogenic factors, contributing to improved tissue repair and regeneration [85]. These advantages position 3D culture platforms as a superior approach for stem cell expansion, therapeutic applications, and regenerative medicine research compared to conventional 2D systems.

Conventional 2D culture methods for EV production are constrained by the limited surface area of culture plates, restricting EV yield. Furthermore, monolayer cultures reduce cell–cell interactions and alter EV composition and function. In contrast, 3D culture systems support the expansion of large cell populations in a time- and cost-efficient manner while preserving cellular phenotypes. Moreover, 3D culture platforms improve EV delivery, enabling sustained release and enhancing cell retention time [9].

Organoid technology offers promising avenues for clinical advancement by bridging the gap between preclinical animal studies and clinical trials. Patient-derived organoids facilitate personalized drug screening, allowing for the identification of optimal treatments [86]. Their scalability in microplate formats supports automated, high-capacity therapeutic screening, while their sustained viability permits the study of long-term drug effects, toxicity, and resistance [87,88]. Ultimately, organoid transplantation holds significant potential—autologous organoids minimize immune rejection and offer regenerative capabilities. As the field advances, organoid-based therapies may revolutionize the treatment of organ failure, cancer, and wound healing.

The 3D structure of organoids closely mimics the native tissue microenvironment, resulting in EVs that more accurately reflect in vivo conditions. OEVs are nanosized lipid vesicles released by organoids and provide robust therapeutic capacities for a range of diseases. Like conventional EVs, OEVs are enriched with biomolecules that regulate gene expression in recipient cells and offer several advantages, including low immunogenicity, effective cargo transport, and reduced risks associated with whole cell-based therapies [54]. Notably, OEVs demonstrate higher production rates and improved physiological functionality, making them highly adaptable for therapeutic engineering. OEVs also exhibit greater tissue specificity and molecular complexity than 2D-EVs and 3D-EVs [89].

Retinal OEVs have been shown to enhance corneal epithelial wound healing, highlighting their therapeutic potential in ocular diseases [77]. This functional specificity could enable more targeted and effective therapies. Moreover, the unique protein and RNA profiles of OEVs make them valuable for biomarker discovery, particularly in cancer research. The tissue-specific nature of organoids enables identification of more precise and clinically relevant biomarkers compared to 2D or conventional 3D culture systems [90].

Despite these advantages, OEVs face several challenges in clinical application, including the need for standardized isolation and characterization methods and GMP-compliant, large-scale production protocols [91]. The quality and quantity of organoids may influence both the yield and therapeutic capacity of the resulting OEVs. Many organoids lack key cellular components such as stromal, immune, or vascular cells, limiting their capacity to model complex tissue interactions. For example, tumor organoids often lack immune cells, which are essential for studying EV-mediated immunomodulation. Additionally, the diverse array of growth factors and cytokines used in organoid culture can influence EV cargo composition, potentially impacting reproducibility and consistency across studies. The isolation and purification of OEVs present significant technical hurdles, especially when organoids are embedded in synthetic scaffolds such as hydrogels or ECM components. Further research is also required to elucidate the molecular mechanisms underlying OEV function and to enhance their therapeutic potential. Fig. 3 summarizes key hurdles in the development and application of OEVs.

To overcome these limitations, future research should prioritize the development of efficient and reproducible OEV isolation techniques, such as microfluidic sorting, to preserve vesicle integrity. Establishing standardized protocols for organoid culture and OEV characterization is critical for reducing variability and improving reproducibility. Incorporating essential cellular components—such as stromal and immune cells—into organoid systems may better recapitulate the native tissue microenvironment and enhance biological relevance. In addition, bioengineering approaches should be leveraged to functionalize OEVs for targeted drug delivery. In summary, while organoids represent a promising platform for EV research, addressing these technical and biological challenges is crucial to unlocking their full diagnostic and therapeutic potential. Future in-depth studies are needed to address these challenges and realize the clinical promise of OEVs.