Establishing a Working Range for Effective MSC-EV Dose

Listen to this Blog:

What is a Therapeutic Dose for EVs?

Extracellular vesicles derived from MSCs (MSC-EVs) are promising therapeutic candidates that have been used in several clinical trials for indications including graft versus host disease (GvHD), kidney disease, and diabetes, among many others [1]. The number of clinical trials investigating MSC-EVs as therapeutics has increased dramatically from 2006 (2 trials) to 2020 (74 trials). To learn more about MSC-EVs journey as therapeutic, check out a recent RoosterBio blog post.

The growing number of clinical trials demonstrates the exciting therapeutic promise of MSC-EVs. However, the dose of MSC-EVs applied to patients across these studies has varied considerably in terms of the method of MSC-EV quantification and the amount of applied MSC-EVs. A single dose is a unit and amount of a therapeutic compound to be applied to a patient. Experimental medicine that occurs in clinical trials typically involves dose regimens, which include multiple doses of the therapeutic compound over a period of time, with each dose occurring at a specific frequency [2]. The total dose is the sum of all doses in a dose regimen. Being novel therapeutic candidates, given their heterogeneity and often limited in vivo retention, MSC-EVs challenge the conventional dose regimen framework. Consensus MSC-EV dose definitions are still in development across scientific and medical fields. In this blog article, we will explore the current state of MSC-EV dosing in completed and ongoing clinical trials, discuss the factors that affect MSC-EV dose, including MSC-EV quantification methods, and establish a target ‘working range’ for MSC-EV dose in humans.

The current consensus affirms that EVs are small particles 50-1000 nm in diameter, comprised of a lipid bilayer containing a variety of biological cargoes, including nucleic acids, proteins, and lipids. Thus, an ‘amount’ of EVs in a preparation can be determined by either quantifying the number of EVs in the preparation (using nanoparticle tracking analysis) or by quantifying the number of nucleic acids, proteins, or lipids (using colorimetric or fluorometric assays) in the preparation. An alternative, simpler yet indirect method to estimate the amount of EVs in a preparation is by counting the number of cells cultured in a given medium to produce EVs. Crucially, any of these three methods can be used in a dose escalation/response study for researchers to claim that a specific unit and amount of EVs achieved the desired therapeutic effect. However, although used previously, the cell counting method does not provide a quantitative measurement of EVs and is thus not recommended for industry applications.

To date, MSC-EV doses in clinical trials have been reported using all three above methods. However, these methods are non-standardized, which can lead to variability across trials. In one clinical study, conditioned medium containing MSC-EVs derived from 40 million cells was administered intravenously to a patient every 2-3 days, totaling 4 doses per regimen, and thus requiring 160M cells in total [3]. Another clinical study used intravenous administration of 100 µg MSC-EV protein per kg patient body weight [4], where protein content is measured using Bradford assay. In a more recent trial targeting pneumonia resulting from COVID-19 infection, patients inhaled 1.25×1010 EVs twice per day over 10 days, totaling 2.5×1011 EVs per dose regimen [5]. These studies differ dramatically by route of administration, EV quantification method, and dose regimen, demonstrating the variation in EV administration in humans on a trial-by-trial basis, which motivates the need to better standardize reported EV dose descriptions.

Ultimately, the effective EV dose will depend on the specific disease indication. However, determining effective EV dose stands to benefit from ongoing work that can provide as much information as possible for researchers to understand typical quantities of EVs being administered to patients – a ‘working range’ of EV dose in humans. To evaluate this ‘working range’ of EV dose, we must consider the administration route, the EV quantification method, and the number of doses occurring per dose regimen.

How Does Administration Route Affect Dose?

For novel therapeutics, the route of administration is important since it directly precedes how therapeutics interact with the body; a concept known broadly as pharmacokinetics. This field uses ADME (Absorption, Distribution, Metabolism, Elimination) to describe the process that occurs after therapeutic administration [6].

  • Absorption: Different classes of therapeutics show different absorption profiles depending on administration route, and thus some administration routes for certain therapeutics require a significantly lower or higher dose [6].
  • Distribution: Distribution of EVs throughout the body is reported to be heterogeneous and depends on parent cell type, source, EV composition, and administration route [7]. Recent studies have used sophisticated biological labeling strategies to track EV distribution in animals [8], likely to be translated to humans in the near future.
  • Metabolism: Whether EVs are metabolized by specific organs or within the bloodstream is likely to depend on administration route [9]. Some evidence shows that patrolling macrophages in the bloodstream take up EVs efficiently in zebrafish [10], but this is yet to be observed in humans.
  • Elimination: Many studies report that EVs are cleared rapidly (within hours) after administration [11]. These results will drive clinical work into establishing prolonged dosing regimens involving multiple doses over time.

In summary, administration route is likely to affect the effective EV dose, and thus should be considered carefully. Most preclinical studies in animals used intravenous administration, direct (site-specific) injection, or subcutaneous injection [12].

The EV Quantification Challenge

Early reporting of EV quantity in preclinical studies focused on the number of cells used in producing a given quantity of medium over a given time interval. As preclinical studies evaluating EVs commenced in animal models, investigators began reporting EV quantity in terms of protein or lipid content, a method typically used for other preclinical biologic compounds. Though this method is acceptable, it is problematic because EVs are not made up entirely of protein or lipid, as they are a mix of several biological components. This is further complicated by the fact that non-EV-specific proteins and lipids exist within the conditioned medium, either in a soluble form or in complex with the EV surface. These contaminants require purification to remove, and studies sometimes do not report or otherwise incompletely define the purification methods used. Thus, reports of EV quantity in terms of total protein or lipid content are dependent on many factors and are unlikely to be comparable across other studies without additional information.

To potentially solve this problem, nanoparticle tracking analysis (NTA) was developed and used [13] to measure the number of EVs and their size distribution. While this information is critical, EVs can range variably in size between 50-1000 nm in diameter. It stands to reason, for example, that larger EVs probably contain more biological material, and thus the contents within larger EVs are likely overrepresented for the total sample. Furthermore, the components within EVs mediating therapeutic effects are biological cargoes, unless EVs are loaded artificially with small molecule drugs or metabolites [14]. Information about these cargoes is not necessarily contained within the data for number of vesicles or their size distribution. Thus, the combination of measurements for EV number, size, and content is necessary to best describe an EV population for dosing purposes.

To attempt to resolve the above issues, the International Society of Extracellular Vesicles (ISEV) lays out the minimal information required for studies of extracellular vesicles (MISEV) which includes the recommendation above. According to MISEV, a description of a given EV composition should include the parent cell number and source, the EV number and size, and the EV protein content, lipid content, RNA content, and presence of surface markers on EVs [15]. In the near future, better technologies such as flow cytometry or single particle interferometric imaging will be utilized to characterize individual EVs within a population to better describe the entire population [16]. However, as it stands currently, dosing information remains varied across studies, making it necessary to evaluate studies individually to make more general conclusions.

Establishing a Working Range for Effective EV Dose

In general, preclinical work done in rodents has typically reported MSC-EV dose in terms of micrograms (µg) of protein [12]. After reviewing 18 preclinical studies* performed across different disease models in rodents, it was found that MSC-EV dose ranges between 0.10 and 250 µg of protein with a mean value of 70 µg. Assuming 1 µg equates to around 2 x 109 MSC-EVs [17], this yields a rough estimate of 1.4 x 1010 MSC-EVs per dose being used on average in preclinical rodent studies. Given that humans are much larger than rodents, it is probably safe to assume that a dose of 1.4 x 1010 MSC-EVs is close to an absolute minimum dose for humans [18]. As pointed out by Sverdlov, the basal concentration of EVs in human blood is roughly 1.2 x 109 per mL [17]. Assuming the average human has 5L of blood in circulation (according to, this means there are roughly 6 x 1012 EVs circulating in the average human bloodstream at any given time. It seems unreasonable to supply a patient with more EVs than currently exist in their bloodstream, so this value is probably a safe upper limit. Thus, a working total MSC-EV dose likely falls somewhere within the broader range of 1 x 1010-6 x 1012 MSC-EVs.

Identifier Indication Admin. Route Doses Interval Per Dose Total
NCT04388982 (1) Alzheimer’s Disease nasal drip 24 twice weekly 5-20 µg

(1-4 x 109)

120-480 µg

(2.4-9.6 x 1011)

NCT04657458 (2) ARDS Intravenous 1 N/A 8 x 1011 8 x 1011
NCT04493242 (3) ARDS Intravenous 1 N/A 0.008-1.2 x 1012 0.008-1.2 x 1012
NCT04602104 (4) ARDS Inhalation 7 daily 0.2-1.6 x 109 0.14-1.2 x 1010
NCT04213248 (5) cGVHD Eye drop 56 4x daily over 14 days 10 µg

(2 x 1010)

560 µg

(1.1 x 1012)

In Human Study (6) Chronic Kidney Disease Intravenous 2 weekly 6.6 mg

(1.3 x 1013)

12.2 mg

(2.6 x 1013)

NCT04202770 (7) Dementia Intravenous (Drip) 1 N/A eq. to 21M cells eq. to 21M cells
NCT04313647 (8) General Safety Inhalation 1 N/A 0.2-1.6 x 109 0.2-1.6 x 109
In Human Study (9) GvHD Intravenous 4 every 2-3 days eq. to 40M cells eq. to 160M cells
In Human Study (10) Hearing loss Intracochlear 1 N/A 3 x 109 3 x 109
NCT03437759 (11) Macular Holes Topical (Drip) 1 N/A 20-50 µg

(0.4-1 x1011)

20-50 µg

(0.4-1 x1011)

NCT04356300 (12) Multiple organ dysfunction Intravenous 14 daily 150mg

(3 x 1014)


(4.2 x 1015)

NCT05060107 (13) Osteoarthritis Local injection 1 N/A 3-5 x 1011 3-5 x 1011
NCT04544215 (14) Pulmonary infection Inhalation 7 daily 0.8-1.6 x 109 0.56-1.1 x 1010
NCT04276987 (15) Severe COVID-19 Inhalation 5 daily 2 x 108 1 x 109
NCT04798716 (16) Severe COVID-19 Intravenous 3 every other day 2-8 x 109 1.4-2.4 x 1010
NCT04491240 (17) Severe COVID-19 Inhalation 20 twice daily over 10 days 0.5-2 x 1010 1-4 x 1010
NCT02138331 (18) Type I Diabetes Mellitus Intravenous 2 weekly 7.5-9.4 x 107 1.5-1.9 x 108

Table 1**. Therapeutic EV studies in humans reporting dose. Search used as of 09Nov2021 with terms “MSC extracellular vesicles” or “MSC exosomes”. Protein content to EV number conversion assumes 2 x 109 MSC-EVs per 1 µg protein. eq. = equivalent. For (6) and (14), assuming an average adult human measures 66 kg.

After performing a search of human studies involving MSC-EVs as the therapeutic compound, 23 human studies and clinical trials were found with only 18 of them reporting the MSC-EV dose used (Table 1**). Of these studies, 11 report dose based on MSC-EV number, 5 report dose based on protein content, and 2 report dose based on cell number. Notably, 44% of studies use intravenous administration and 36% use inhalation, with the remainder using topical or local injection. Furthermore, 11 of 18 (61%) use a dose regimen versus a single dose. Excluding the two studies reporting cell number, the total MSC-EV dose ranges from 1.7 x 108 to 4.2 x 1015, with a median value of 4.3 x 1010. Importantly, this value exists squarely within the broader range determined above, indicating an agreement between this theoretical prediction and the reality occurring in ongoing clinical trials. Thus, the working range of EV dose exists within 1 x 1010-6 x 1012 with a target dose around 4 x 1010.

Nailing the Number: Large-scale EV Production

Now that a working range for EV doses in humans has been established, an important question arises — how feasibly can batches of EVs be produced to this scale? Though total dose will vary by indication, given a conservative estimate of 1000 EVs produced per cell per production process [17,19], it is likely that 100-1000M MSCs will be required to create a single dose regimen in the working range, which is in line with predictions reported previously [20]. To meet the growing demand for MSC-EV therapeutic investigations, existing manufacturing processes will require significant improvements. Furthermore, since EV properties can differ according to production conditions, it is imperative to develop scalable methods for EV production to ensure consistent potency in the desired therapeutic indication before a significant investment in clinical translation of EV products. Our goal at RoosterBio is to be at the forefront of establishing processes and products that seek to maximize productivity of hMSCs, and the EVs that they produce, to meet this growing demand.  We will address EV manufacturing processes and variables that drive productivity and yield in future blog posts.

  1. Wiklander, O. P. B., Brennan, M. Á., Lötvall, J., Breakefield, X. O. & El Andaloussi, S. Advances in therapeutic applications of extracellular vesicles. Science Translational Medicine 11, eaav8521, doi:10.1126/scitranslmed.aav8521 (2019).
  2. Grudzinskas, C. in Principles of Clinical Pharmacology (Second Edition)   (eds Arthur J. Atkinson et al.)  501-517 (Academic Press, 2007).
  3. Kordelas, L. et al. MSC-derived exosomes: a novel tool to treat therapy-refractory graft-versus-host disease. Leukemia 28, 970-973, doi:10.1038/leu.2014.41 (2014).
  4. Nassar, W. et al. Umbilical cord mesenchymal stem cells derived extracellular vesicles can safely ameliorate the progression of chronic kidney diseases. Biomater Res 20, 21-21, doi:10.1186/s40824-016-0068-0 (2016).
  5. Evaluation of Safety and Efficiency of Method of Exosome Inhalation in SARS-CoV-2 Associated Pneumonia,
  6. Eddershaw, P. J., Beresford, A. P. & Bayliss, M. K. ADME/PK as part of a rational approach to drug discovery. Drug Discovery Today 5, 409-414, doi: (2000).
  7. Wiklander, O. P. B. et al. Extracellular vesicle in vivo biodistribution is determined by cell source, route of administration and targeting. Journal of extracellular vesicles 4, 26316-26316, doi:10.3402/jev.v4.26316 (2015).
  8. Verweij, F. J. et al. The power of imaging to understand extracellular vesicle biology in vivo. Nature Methods 18, 1013-1026, doi:10.1038/s41592-021-01206-3 (2021).
  9. Mead, B. & Tomarev, S. Bone marrow‐derived mesenchymal stem cells‐derived exosomes promote survival of retinal ganglion cells through miRNA‐dependent mechanisms. Stem cells translational medicine 6, 1273-1285 (2017).
  10. Verweij, F. J., Hyenne, V., Van Niel, G. & Goetz, J. G. Extracellular Vesicles: Catching the Light in Zebrafish. Trends in Cell Biology 29, 770-776, doi: (2019).
  11. Lai, C. P. et al. Dynamic Biodistribution of Extracellular Vesicles in Vivo Using a Multimodal Imaging Reporter. ACS Nano 8, 483-494, doi:10.1021/nn404945r (2014).
  12. Tieu, A. et al. An Analysis of Mesenchymal Stem Cell-Derived Extracellular Vesicles for Preclinical Use. ACS Nano 14, 9728-9743, doi:10.1021/acsnano.0c01363 (2020).
  13. Sokolova, V. et al. Characterisation of exosomes derived from human cells by nanoparticle tracking analysis and scanning electron microscopy. Colloids and Surfaces B: Biointerfaces 87, 146-150, doi: (2011).
  14. Herrmann, I. K., Wood, M. J. A. & Fuhrmann, G. Extracellular vesicles as a next-generation drug delivery platform. Nature Nanotechnology 16, 748-759, doi:10.1038/s41565-021-00931-2 (2021).
  15. Théry, C. et al. 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. J Extracell Vesicles 7, 1535750, doi:10.1080/20013078.2018.1535750 (2018).
  16. Bordanaba-Florit, G., Royo, F., Kruglik, S. G. & Falcón-Pérez, J. M. Using single-vesicle technologies to unravel the heterogeneity of extracellular vesicles. Nature Protocols 16, 3163-3185, doi:10.1038/s41596-021-00551-z (2021).
  17. Sverdlov, E. D. Amedeo Avogadro’s cry: What is 1 µg of exosomes? BioEssays 34, 873-875, doi: (2012).
  18. Nair, A. B. & Jacob, S. A simple practice guide for dose conversion between animals and human. J Basic Clin Pharm 7, 27-31, doi:10.4103/0976-0105.177703 (2016).
  19. Record, M., Subra, C., Silvente-Poirot, S. & Poirot, M. Exosomes as intercellular signalosomes and pharmacological effectors. Biochemical Pharmacology 81, 1171-1182, doi:10.1016/j.bcp.2011.02.011 (2011).
  20. Olsen, T. R., Ng, K. S., Lock, L. T., Ahsan, T. & Rowley, J. A. Peak MSC—Are We There Yet? Frontiers in Medicine 5, doi:10.3389/fmed.2018.00178 (2018).
* Studies used in this calculation:
  1. Aliotta, J. M. et al. Exosomes induce and reverse monocrotaline-induced pulmonary hypertension in mice. Cardiovascular Research 110, 319-330, doi:10.1093/cvr/cvw054 (2016).
  2. Bian, S. et al. Extracellular vesicles derived from human bone marrow mesenchymal stem cells promote angiogenesis in a rat myocardial infarction model. Journal of Molecular Medicine 92, 387-397, doi:10.1007/s00109-013-1110-5 (2014).
  3. Bruno, S. et al. Microvesicles derived from mesenchymal stem cells enhance survival in a lethal model of acute kidney injury. PloS one 7, e33115 (2012).
  4. Fang, S. et al. Umbilical cord‐derived mesenchymal stem cell‐derived exosomal microRNAs suppress myofibroblast differentiation by inhibiting the transforming growth factor‐β/SMAD2 pathway during wound healing. Stem cells translational medicine 5, 1425-1439 (2016).
  5. Lai, R. C. et al. Exosome secreted by MSC reduces myocardial ischemia/reperfusion injury. Stem cell research 4, 214-222 (2010).
  6. Lee, C. et al. Exosomes Mediate the Cytoprotective Action of Mesenchymal Stromal Cells on Hypoxia-Induced Pulmonary Hypertension. Circulation 126, 2601-2611, doi:10.1161/CIRCULATIONAHA.112.114173 (2012).
  7. Li, T. et al. Exosomes derived from human umbilical cord mesenchymal stem cells alleviate liver fibrosis. Stem cells and development 22, 845-854 (2013).
  8. Mead, B. & Tomarev, S. Bone marrow‐derived mesenchymal stem cells‐derived exosomes promote survival of retinal ganglion cells through miRNA‐dependent mechanisms. Stem cells translational medicine 6, 1273-1285 (2017).
  9. Monsel, A. et al. Therapeutic Effects of Human Mesenchymal Stem Cell–derived Microvesicles in Severe Pneumonia in Mice. American Journal of Respiratory and Critical Care Medicine 192, 324-336, doi:10.1164/rccm.201410-1765OC (2015).
  10. Phinney, D. G. et al. Mesenchymal stem cells use extracellular vesicles to outsource mitophagy and shuttle microRNAs. Nature Communications 6, 8472, doi:10.1038/ncomms9472 (2015).
  11. Sicco, C. L. et al. Mesenchymal stem cell‐derived extracellular vesicles as mediators of anti‐inflammatory effects: Endorsement of macrophage polarization. Stem cells translational medicine 6, 1018-1028 (2017).
  12. Tan, C. Y. et al. Mesenchymal stem cell-derived exosomes promote hepatic regeneration in drug-induced liver injury models. Stem cell research & therapy 5, 1-14 (2014).
  13. Teng, X. et al. Mesenchymal Stem Cell-Derived Exosomes Improve the Microenvironment of Infarcted Myocardium Contributing to Angiogenesis and Anti-Inflammation. Cellular Physiology and Biochemistry 37, 2415-2424, doi:10.1159/000438594 (2015).
  14. Xin, H. et al. Systemic administration of exosomes released from mesenchymal stromal cells promote functional recovery and neurovascular plasticity after stroke in rats. Journal of Cerebral Blood Flow & Metabolism 33, 1711-1715 (2013).
  15. Yu, B. et al. Exosomes secreted from GATA-4 overexpressing mesenchymal stem cells serve as a reservoir of anti-apoptotic microRNAs for cardioprotection. International Journal of Cardiology 182, 349-360, doi: 10.1016/j.ijcard.2014.12.043 (2015).
  16. Zhang, J. et al. Exosomes released from human induced pluripotent stem cells-derived MSCs facilitate cutaneous wound healing by promoting collagen synthesis and angiogenesis. Journal of translational medicine 13, 1-14 (2015).
  17. Zhu, Y.-g. et al. Human Mesenchymal Stem Cell Microvesicles for Treatment of Escherichia coli Endotoxin-Induced Acute Lung Injury in Mice. STEM CELLS 32, 116-125, doi:10.1002/stem.1504 (2014).
  18. Zou, X. et al. Microvesicles derived from human Wharton’s Jelly mesenchymal stromal cells ameliorate renal ischemia-reperfusion injury in rats by suppressing CX3CL1. Stem cell research & therapy 5, 1-13 (2014).
**Studies reporting EV dose:
Studies not reporting EV dose:

We're fueling the rapid commercialization of scalable regenerative cures. If you're interested in having a conversation about how RoosterBio can accelerate your product & process development and shorten your time to the clinic, contact us.