Products Why TechnoCell Support Facility Safety & Quality Technical Notes News Membership
Member Login Contact

Technical Notes

Can You Trust “X Trillion Particles”?
How to Read an Exosome Count, and Three Traps

“X trillion high-concentration exosomes per vial.” Claims of this kind turn up more and more often in product brochures and on websites. The international guidelines for extracellular vesicle research (MISEV2023), however, warn in plain terms against taking such particle counts at face value. Why is the number in the brochure unreliable, and what should you look at instead to judge the quality of a preparation? We work through it with as little jargon as possible.

This article in 30 seconds

  • The “count” moves a millionfold with the method. Comparing counts across products measured on different instruments tells you nothing.
  • Impurities are counted along with the vesicles. A document that says “X exosomes” may in fact be counting clumps of protein as well.
  • More is not stronger. As the science stands, there is no basis for reading a larger count as a larger effect.

Below we take these three points in turn, as three traps. Every one of them can be checked against the original wording of the international guidelines.

TRAP 01

No instrument can yet count “exosomes” accurately

Why “X trillion particles” is not a reliable figure

Research on extracellular vesicles (the tiny membrane-bound sacs that cells release; exosomes are one kind) follows reporting guidelines issued by the International Society for Extracellular Vesicles (ISEV). The current version is MISEV2023[1].

Those guidelines state explicitly that measuring particle counts has limits. The reason is simple: no instrument available today can single out exosomes and count them exhaustively.

“…the number concentration (in particles/mL), a metric that is widely reported and used for … in vivo dosing. However, it is often unreliable, since many techniques lack specificity for EVs and sensitivity for all EVs.”

In plain terms: combined with a volume measurement, the number of vesicles gives a number concentration (particles/mL) — a metric widely reported and used to standardize assay inputs, to express results, and to set doses in vivo. Yet it is often unreliable, because many techniques are neither specific to extracellular vesicles nor sensitive enough to capture all of them.

MISEV2023, Section 5.1, “Quantification of particle number concentration”

Which is why the guidelines say not to write “exosome concentration”

MISEV2023 goes further and makes a recommendation.

“Unless methods are highly specific for EVs, the output of these measures should be described as pertaining to ‘particles’ or ‘EPs’.”

In plain terms: unless a method is highly specific for extracellular vesicles, what it outputs should be described as “particles” or “extracellular particles” — not as vesicles.

MISEV2023, Section 5.1, recommendations

In other words, the right name for it is “particle concentration” — a figure that includes impurities of unknown identity. The same goes for size: the guidelines ask for “particle diameter” rather than “vesicle diameter.”

Turn that around, and the moment a document asserts “X exosomes,” you need to establish what that number counted, and how.

TRAP 02

Change the method and the same sample can differ by up to a millionfold

“Supplier A says 10 billion, Supplier B says 10 trillion, so B is better” does not hold

This is a common misreading. For one and the same sample, changing nothing but the instrument (the method) has been reported to change the result by as much as six orders of magnitude — roughly a millionfold.

“Note that reported EV concentration in blood plasma spans six orders of magnitude depending on the measurement method.”

In plain terms: published figures for the concentration of extracellular vesicles in blood plasma span six orders of magnitude — roughly a millionfold — depending on which measurement method was used.

MISEV2023, Section 5.1 (original source: Johnsen et al., 2019)

This is not measurement error. The gap arises because each technique is looking at something different. Lining up figures obtained by different methods and comparing them therefore means nothing.

What each method misses, and what it over-counts

NTA (nanoparticle tracking analysis)
The most widely used method. It illuminates particles and tracks their motion — but in this scheme exosomes cannot be told apart from mere clumps of protein (aggregates). The same holds for serum-derived lipoprotein particles. Whatever is mixed in adds to the count (over-counting). In the other direction, the effective lower limit of detection is reported at roughly 50–70 nm, and 70–90 nm for vesicles: anything smaller is there in the sample but never appears in the count (under-counting).

TRPS (tunable resistive pulse sensing)
Measures the change in electrical resistance as particles pass through a fine pore. The lower limit is set by instrument sensitivity, the upper limit by pore diameter. It is considered strong on resolution in samples with a wide spread of sizes, though NTA has been reported to detect more particles below 150 nm.

ELISA (CD9, CD63, CD81)
Measures the amount of the marker proteins that sit on the vesicle surface. But not every vesicle carries those markers. MISEV2023 states explicitly that methods using these three are not specific to the exosome subtype.

Total protein
The total amount of protein in the sample. Because it includes components derived from the culture medium, it is not a measure of the vesicles themselves.

Put into a single figure, all of this looks like the following.

The particle-size axis and detection limits Extracellular vesicles are distributed mostly between about 30 and 200 nanometers. Lipoprotein particles and protein aggregates exist in the same size range and are counted without being distinguished. The effective lower detection limit of NTA is roughly 50 to 70 nanometers, so vesicles smaller than that never appear in the count. 10nm 50nm 100nm 200nm 1000nm Extracellular vesicles Distributed mostly around 30–200 nm Different things, same size Lipoprotein particles, protein aggregates Effective NTA detection limit (approx. 50–70 nm) Never counted Under-counting and over-counting happen at the same time. Vesicles smaller than the dashed line are present but never counted (under-counting). Lipoproteins and aggregates of the same size are counted as vesicles (over-counting). The particle-size axis and detection limits Extracellular vesicles are distributed mostly between about 30 and 200 nanometers. Lipoprotein particles and protein aggregates exist in the same size range and are counted without being distinguished. The effective lower detection limit of NTA is roughly 50 to 70 nanometers, so vesicles smaller than that never appear in the count. 10nm 50nm 100nm 200nm 1000nm Extracellular vesicles Mostly 30–200 nm Different things, same size Lipoproteins, aggregates NTA detection limit (approx. 50–70 nm) Under-counting and over-counting, at once. Vesicles below the dashed line are never counted. Lipoproteins and aggregates of the same size are counted as if they were vesicles.
Figure 1: The particle-size axis. Anything below the detection limit goes uncounted, while different things of the same size are counted together.

On the mode diameter NTA reports for particles of low refractive index, MISEV2023 cautions that it “may reflect the instrument’s limit of detection” rather than the true mode of the vesicle population, and asks for the size distribution itself to be shown, not only an average. A document that carries a single “mean X nm” and nothing else is, on this point, short of information.

Nor do the guidelines regard any single method as sufficient.

“As a result, no single measurement or method is able to satisfy all EV characterization requirements, and use of orthogonal methods (those that do not have the same measurement limitations) is recommended.”

In plain terms: no single measurement or method can satisfy every requirement for characterizing extracellular vesicles, so combining orthogonal methods — methods that do not share the same measurement limitations — is recommended.

MISEV2023, opening of Section 5, “Characterization of EVs”
TRAP 03

“More particles means more effect” has, as yet, no scientific basis

This is the most important point

As things stand, there is no evidence that a higher particle count produces a larger biological effect.

On the contrary, there are reports of preparations whose particle counts, particle sizes, and levels of representative markers were comparable, yet whose actual biological activity was entirely different. With no standardized potency assay (a test of how strongly something acts) established, no criterion of the form “this many particles is enough to work” exists — that is, no such figure as an optimal particle count exists at all.

The quantities used in clinical studies likewise vary from one preparation method to the next, and are not in a form that allows cross-comparison.

Comparing products by the size of the “count” in a brochure is, at present, a reading with nothing behind it.

To avoid a misunderstanding: measuring particle counts is not itself pointless. Measured consistently by the same method, the count works well as an indicator of lot-to-lot variation. The problems are two: ranking products by figures obtained with different methods, and reading a larger count as a larger effect.

Four things to check when reading product literature

So when documents from several manufacturers are put in front of you, where should you look to get closer to real quality? These four points.

  1. Does it state which instrument (method) was used?

    A figure with no method stated cannot meaningfully be compared with anyone else’s. As above, the instrument alone can account for a difference of up to a millionfold.

  2. Is protein amount given alongside the particle count?

    Guidance from the Japanese Society for Regenerative Medicine[2] considers it desirable for particle count and protein amount to correlate. One figure on its own does not tell you whether what is in those particles is dilute or concentrated.

  3. How many surface markers, and analyzed how?

    An international position paper calls for semi-quantitative analysis of at least three markers. A single marker makes for weak evidence.

  4. How is non-vesicular material assessed and removed?

    A large count is worth little if most of it is unwanted protein aggregate. MISEV2023 lists verifying how much non-vesicular material is present among its reporting items.

A document that covers all four can at least be read as putting its figures forward with the limits of measurement understood. A document where only “X trillion particles” is set in large type holds no more information than that.

A note on terminology — “exosome” and “extracellular vesicle (EV)”

MISEV asks that the generic term extracellular vesicle (EV) be used unless the biogenesis pathway can be strictly demonstrated, because “exosome” implies origin from an intracellular structure called the multivesicular body.

Most commercial products are labeled “exosomes,” but in practice they are usually a mixture of vesicles whose origin has not been determined.

A document that deliberately keeps the precise terms apart — “extracellular vesicles (EVs),” “secretome” — is one sign of a manufacturer working from the academic background.

The definition of exosomes themselves, how they are formed, and their structure are covered in What Is an Exosome; what is lost when only the membrane sacs are isolated, in What Is Lost When Only the Exosomes Are Isolated; and the specification items required by Japanese guidance, in What Is Stem Cell Conditioned Medium.

Frequently Asked Questions

Is it enough to choose the product with the highest particle count?

There is currently no evidence to support that judgment. Preparations with comparable particle counts, particle sizes, and levels of representative markers have been reported to differ substantially in biological activity. And because no standardized potency assay has been established, no such figure as an optimal particle count exists. Rather than the size of the number, check which method was used and how non-vesicular contaminants were assessed.

Can your figures be compared with another supplier’s particle counts?

Not unless the measurement method is the same. MISEV2023 notes that reported concentrations of extracellular vesicles in plasma span six orders of magnitude (roughly a millionfold) depending on the method. This is not measurement error; it arises because each technique sees something different.

Does detection of markers such as CD9 prove high quality?

Markers corroborate that vesicles are present; they are not an index of quality. Not all vesicles carry the same markers, and MISEV2023 states explicitly that methods using CD9, CD63, and CD81 are not specific to the exosome subtype. The same guidelines also state that no universal marker able to identify all extracellular vesicles, whatever their origin, is currently known.

References

  1. [1] Welsh JA, et al. Minimal information for studies of extracellular vesicles (MISEV2023): From basic to advanced approaches. J Extracell Vesicles. 2024;13(2):e12404. (International Society for Extracellular Vesicles, ISEV)
  2. [2] Japanese Society for Regenerative Medicine (日本再生医療学会), in cooperation with the Japanese Society for Extracellular Vesicles (日本細胞外小胞学会): Guidance on the Clinical Application of Extracellular Vesicles and Related Products, 1st Edition, April 30, 2024

The English quotations are excerpted from the original text of MISEV2023; the plain-language restatements are ours.

This article explains measurement methods and their limits. Our stem cell secretome (the totality of components cells secrete during culture) is supplied for research use only and is not intended to demonstrate efficacy against any specific disease.

View Stem Cell Secretome specifications Request analytical data