Data were collected from the institutional databases of the Cochrane Register of Studies, ClinicalTrials.gov, and the Chinese Clinical Trial Registry using the search terms “mesenchymal stem cells (MSCs),” “extracellular vesicles (EVs),” and “exosomes (Exos).” The specific search query employed was (“Mesenchymal Stem Cells” OR “MSCs” OR “Mesenchymal Stem Cell”) AND (“Extracellular Vesicles” OR “EVs”) AND (“Exosomes” OR “Exos”). Clinical trials registered from the inception of these databases to 17 February 2024, were included, without restrictions on study start time or research sponsorship.

The data were downloaded into CSV files and included the following items: registration number, trial title, recruitment status, sponsor, clinical stage, country of origin, and registration date. Additional information, such as disease targets, cell sources, routes of administration, and doses, was manually extracted from the trial registration records when it could not be directly matched and downloaded from the databases. Due to incomplete data in many categories, a different number of trials may be included in each parameter analyzed. Duplicate clinical trials were identified and eliminated based on registration numbers before entering the subsequent analysis process (Table 1).

The inclusion criteria for clinical trials in the search results are:

The exclusion criteria for clinical trials are:

The screening process is illustrated in Figure 1.

The clinical trial research data collection was conducted by two trained researchers. After retrieving all clinical trials, two independent researchers screened the included trials based on the duplicate data and the inclusion and exclusion criteria. Controversial clinical trial research was referred to a third trained researcher for resolution. This process ensured the final determination of the clinical trial projects included in this systematic visualization analysis.

Three independent researchers evaluated all clinical trial titles and research statuses. Controversial parts were resolved through communication and discussion among the three researchers. The extraction of relevant information from the final included clinical trial studies encompassed: clinical trial registration number, study start year, country and unit, study type, study phase, study design, route of administration, disease and injury type, source of exosomes or extracellular vesicles, drug dosage description or treatment regimen, publication of study results and Study URL. The latest dataset is generated in the Supplementary Appendix Tables 1, 2.

Power BI 2.1 software was used to generate datasets based on the research detailed in Supplementary Appendix Tables 1, 2. These visualized various aspects of clinical trial research, including time, country and region, type, design, stage, indication, and source of MSCs. The research stage classifications were based on the descriptions of clinical trial stages by the China Center for Food and Drug International Exchange (CCFDIE) and the National Medical Products Administration (NMPA), categorizing exploratory trials/preliminary experiments as Phase I/Phase II clinical research.

Clinical trials with reported results were identified, and these results were not limited to publication in registration databases, journal literature, journal news, or open access sources. Clinical trials were further categorized according to the route of administration and the type of injury or disease. For clinical trial studies that published application doses and results, the main components, doses, frequency of administration, and effects of the intervention drugs were statistically analyzed.

Due to differences in the unit modes of MSC-EVs or MSC-Exos dose calculations, a unified standard was adopted wherever possible. Tables 1, 2 were summarized based on tolerance, safety, adverse events, and dose-effect trends to provide a comprehensive evaluation of the dose-effect relationship.

Data Collection: The trial item data were downloaded into CSV files and included registration number, trial title, recruitment status, sponsor, clinical stage, country of origin, and registration date. Additional information about the disease, source of MSCs, route of administration, and dose was manually extracted from the trial registration data, as it could not be directly matched and downloaded from the databases. In many cases, complete data for all these categories were not found, but data were collected whenever possible, resulting in different numbers of trials included in each parameter analyzed.

Screening Process: A total of 187 clinical trials were retrieved. After eliminating 39 duplicate trials and excluding 38 trials through preliminary screening of titles and abstracts, the research data of 90 clinical trials were evaluated against the inclusion criteria. Finally, 66 clinical trials were included in the analysis. The screening process is illustrated in Figure 2, and an overview is shown in Figure 1. The generated dataset is presented in Supplementary Appendix Table 1, which includes the baseline data of clinical trials of MSC-EVs and MSC-Exos, and Supplementary Appendix Table 2, which details the intervention methods of these.

The registration of clinical trials related to MSC-EVs and MSC-Exos has been ongoing from April 2014 to the present. The dataset included in this study covers a span of ten years. The distribution of countries/regions involved in these studies shows that China ranks first with 30 studies, followed by Iran and the United States, each with 12 studies, indicating high research activity. Other countries such as Turkey, Brazil, Germany, and the Russian Federation each have 2 studies, while Singapore, Egypt, Indonesia, Thailand, Italy, Greece, and Israel have only 1 study each, as shown in Figures 3, 4. The number of studies before 2020 was only 5, but 2020 marked a breakthrough, showing an explosive growth trend, including 26 studies on lung diseases and 23 clinical trials related to COVID-19. According to Medidata, the COVID-19 pandemic negatively impacted ongoing and newly launched clinical trials worldwide, particularly causing delays in patient enrollment and recruitment. However, clinical trials related to MSC-EVs and MSC-Exos have shown an upward trend and are actively used to explore treatments for COVID-19. The research phases of related clinical trials are predominantly preclinical, phase I, or early phase I and II studies, with 61 trials (92.24%) falling into these categories (Figure 5). Additionally, interventional studies dominate the research methods of clinical trials related to MSC-EVs and MSC-Exos, with 60 trials (90.91%) adopting this approach. The preferred research designs are parallel trials (33 trials, 50.00%) and single-arm trials (20 trials, 30.30%).

We made extensive efforts to collect the sources of MSCs from study titles, abstracts, designs, and intervention methods in the registration database through the MSC-EVs and MSC-Exos related clinical trial datasets. Despite supplementing missing sources in the records, there were still 22 studies that did not clearly mention the tissue sources of the MSCs used.

The clearly stated tissue sources in the dataset include: Umbilical cord, Bone marrow, Adipose tissue, Placenta, Wharton’s jelly, Pluripotent stem cells, Tumor stroma. Among these, umbilical cord (14 studies, 29.55%), bone marrow (13 studies, 29.55%), and adipose tissue (10 studies, 22.73%) are the most common sources, with all three having conducted phase III clinical trials (Figure 6). This trend may indicate that factors such as the difficulty of obtaining different tissue sources, immunogenicity, and proliferation rate have limited the development of clinical trials related to MSCs from other tissues. Additionally, it suggests that an increasing number of clinical trials are using EVs and Exos extracted from commercially available MSCs cell banks.

Our dataset examined the specificity of diseases in clinical trials, revealing a significant focus on pulmonary disease (26 studies) and a high number of COVID-19-related clinical trials (23 studies). Additionally, skin, eye, anorectal, bone, and joint diseases have also been widely studied.

The most common drug administration methods for lung diseases are: intravenous infusion (15 studies), aerosolization (12 studies). Statistics on the sources of MSCs used in different disease types show that in the treatment of lung diseases, MSC-EVs and MSC-Exos derived from bone marrow (BM) and umbilical cord (UC) tissues are significant research hotspots (Figures 7, 8). This trend highlights the relevance and potential of BM and UC-derived MSC-EVs and MSC-Exos in treating lung diseases, particularly in the context of the COVID-19 pandemic. The detailed analysis of the dataset underscores the evolving focus on specific MSCs sources and administration methods to optimize therapeutic outcomes in various disease contexts.

The main routes of administration for clinical studies of MSC-EVs and MSC-Exos are intravenous infusion, nebulization, and local administration. The dataset includes the following: intravenous injection (24 studies), local administration (15 studies), nebulization (13 studies).

Analysis of the distribution of clinical trial research stages reveals that intravenous infusion and nebulization cover the widest range of clinical trial stages. Both methods have advanced to clinical phase III trials, indicating that intravenous infusion and nebulization are at the forefront of clinical research and are relatively mature methods (see Figure 9). This analysis highlights the prominence and progress of intravenous infusion and nebulization as preferred administration routes in the clinical study of MSC-EVs and MSC-Exos. Their maturity and widespread application across different clinical trial stages underscore their potential for broader clinical adoption and therapeutic use.

This study retrieved data from three major databases, identifying 187 global clinical trials involving MSC-derived EVs and Exos. After applying inclusion and exclusion criteria, 66 relevant trials were included, resulting in two research datasets (see Supplementary Appendix Tables 1, 2). These datasets systematically report on clinical trials related to MSC-EVs and MSC-Exos, with visual analysis highlighting research focus areas across different regions from 2014 to 2024. The analysis reveals preferences in intervention methods (60 trials), phase I/II stages (61 trials), parallel control (33 trials), and single-arm designs (20 trials). Additionally, it identifies research trends in nine administration routes and seven tissue sources. Disease-specific analysis indicates that lung disease (26 trials) is a current research hotspot, with aerosol inhalation being a key administration method. Consistent with the MISEV2018 guidelines published by the International Society for Extracellular Vesicles (ISEV), accurate terminology is essential for the rigorous reporting and interpretation of EV-based research. In this review, we use “extracellular vesicle (EV)” as the generic term and refer to “exosomes” only when the original study provides sufficient evidence of endosomal origin, such as through differential ultracentrifugation, specific marker profiling (e.g., CD63, CD81, CD9), or transmission electron microscopy. Given that most clinical trials do not perform detailed vesicle subtyping, the blanket use of “exosome” in early-phase clinical studies can be misleading. Standardized nomenclature not only supports scientific reproducibility and inter-study comparability but is also critical for regulatory clarity and the development of EV-specific guidelines for clinical use.

Clinical trials with reported results predominantly focus on lung disease, displaying differences in effective dose magnitudes associated with intravenous infusion and aerosol inhalation, as well as a range of minimum effective doses (MEDs) in clinical applications. This dynamic and rapidly evolving field, particularly concerning lung diseases and COVID-19, has significant implications for reducing long-term costs, increasing the reporting of clinical trial results (including negative results), minimizing unnecessary trial duplications, and avoiding ineffective doses.

In-depth exploration of these findings reveals potential challenges and future directions. The rapid response to the COVID-19 pandemic in exploring MSC-EVs and Exos for therapeutic purposes is both timely and crucial (20). These efforts not only address current challenges but also set a precedent for rapid scientific mobilization in future global health crises (21). The preference for intravenous and aerosol routes underscores the strategic focus on maximizing therapeutic efficacy, particularly for respiratory diseases, while local administration may enhance therapeutic effects and minimize systemic exposure (22–24).

The diversity of MSCs sources, particularly umbilical cord, bone marrow, and adipose tissue, highlights the need for a broader understanding of tissue-source specificity and its therapeutic implications (25). Diversifying tissue sources may facilitate more personalized and effective treatments tailored to individual patients or specific disease mechanisms. To fully exploit the therapeutic potential of MSC-EVs and Exos, a deeper understanding of their mechanisms of action is essential (9). The variation in dose calculation units and lack of standardized protocols pose significant challenges, complicating comparisons and meta-analyses. Establishing standardized indicators and ensuring consistency in preparation, characterization, and dose units across different studies is crucial for advancing clinical applications (26–28).

The dose-effect relationship of MSC-EVs and MSC-Exos is vital for optimizing therapeutic outcomes, ensuring safety, and minimizing adverse effects. This relationship is influenced by multiple factors, including the nature of the disease, route of administration, and patient-specific variables. Determining the optimal dose range is critical to maximizing therapeutic effects while minimizing potential risks. Experimental results indicate that the nature of the disease being treated plays a crucial role in determining the appropriate dose. Diseases localized to specific areas, such as wounds or lung diseases, may respond better to lower doses delivered directly to the affected area. In contrast, systemic diseases may require higher doses or targeted delivery strategies to achieve the desired therapeutic effect (29).

For lung diseases, aerosolization can deliver treatment directly to the lungs, potentially requiring lower doses compared to systemic administration routes such as intravenous infusion. This direct delivery can enhance local therapeutic effects while reducing systemic exposure and potential side effects (28, 30–32). The optimal dose range for MSC-EVs and MSC-Exos is narrow, particularly when administered by aerosolization or intravenous routes, emphasizing the importance of precise dosing in clinical applications. Doses that are too low may be ineffective, while doses that are too high may cause adverse reactions. Understanding the biological mechanisms of therapeutic effects can inform more targeted and effective dosing strategies, optimizing doses according to specific therapeutic targets or pathways.

Regulatory and ethical considerations—such as donor variability, batch-to-batch consistency, and long-term safety—are central to the clinical adoption of MSC-EVs and MSC-Exos. Promoting global collaboration and open data sharing can accelerate progress by pooling resources, expertise, and datasets, thereby avoiding duplication and facilitating faster translation into patient benefit. However, commercialization introduces additional challenges, including scaling up production, maintaining product quality, ensuring affordability, and securing equitable access. While commercial cell banks may mitigate some logistical issues, they also raise concerns regarding proprietary technologies, cost structures, and fair distribution.

Despite growing therapeutic promise, MSC-EVs translation remains constrained by persistent methodological, regulatory, and commercial barriers. Unlike conventional biologics or cell-based therapies, MSC-EVs lack standardized definitions for product identity, dosing units, and potency metrics—undermining reproducibility and delaying regulatory approval. Manufacturing heterogeneity, arising from differences in tissue source, donor characteristics, culture conditions, and isolation methods, further contributes to batch inconsistency. The absence of validated potency assays and pharmacokinetic/pharmacodynamic (PK/PD) data limits quality control and leaves uncertainties regarding biodistribution, clearance, and optimal dosing schedules. Cross-trial comparisons are hampered by inconsistent dose reporting (e.g., particle counts, protein content, volume) and reliance on unvalidated surrogate endpoints such as CRP or IL-6. To date, no MSC-EVs product has been approved by the FDA, EMA, or comparable authorities, reflecting a fragmented global regulatory framework for EV classification, GMP compliance, and clinical evaluation.

Compounding these issues is the uncontrolled proliferation of unregulated “exosome therapies” outside formal clinical frameworks, posing substantial risks to patient safety and eroding scientific credibility. Addressing these concerns requires internationally harmonized manufacturing, characterization, and reporting protocols; consensus-driven nomenclature (e.g., per MISEV guidelines); validated clinical endpoints; and regulatory convergence. Designing clinical trials with standardized dosing metrics, robust dose–response evaluation, and comprehensive adverse event reporting will be critical to defining optimal therapeutic windows across diseases and administration routes.

Ultimately, advancing MSC-EVs from experimental tools to evidence-based therapeutics will demand coordinated efforts from academia, industry, and regulatory agencies to establish rigorous standards, protect patient safety, and uphold scientific integrity. As the field evolves, integration of precision medicine approaches and patient-centered strategies will be pivotal to fully realizing the therapeutic potential of MSC-EVs in clinical practice.