Proposal for ISCT to revise the definition and minimal criteria of mesenchymal stromal cells

by Alexey Bersenev on December 14, 2012 · 2 comments

in mesenchymal

Post to Twitter Send Gmail Post to LinkedIn

In 2006 International Society for Cellular Therapy (ISCT) published a position paper on defining a minimal criteria for multipotent mesenchymal stromal cells (MSC). It was very good move! However, since that time, new research data have been published. Some flaws of ISCT definition have been discussed and some statements have been challenged. I think we’re getting to the point when MSC minimal criteria should be revised. I’m starting a discussion with professionals about necessity and content of such revision. Today, I’d like to propose a few positions, which could be considered in revised ISCT statement.

1. The role of ISCT
ISCT is very well respected and trusted organization among professionals. Most of professionals agree with MSC definition, proposed by ISCT. We polled professionals on MSC definition and got the following results:

Therefore, ISCT should continue to take a leading role in discussing and in publishing revised statements.

2. Nomenclature
The current ISCT definition recommends to use term “multipotent mesenchymal stromal cells” (MSC) instead of label “mesenchymal stem cells”. However, my analysis of literature showed that, after 2006 ISCT members (including the authors of MSC position paper) themselves frequently use terms “mesenchymal stromal cells” or “mesenchymal stem cells”. I think, the main point of confusion is a word “multipotent” in ISCT definition. It seem to me that ISCT experts wanted to avoid word “stem” and replaced it by “multipotent”. But “multipotent is actually indicating stem cell nature of the cells. The term “multipotent” was picked based on “tri-lineage differentiation” criteria. But tri-lineage potential is not really reflects multipotency.

I’d propose to remove “multipotent” from revised version and leave “mesenchymal stromal cells” only. Others nomenclature versions should be discussed.

3. Stemness
ISCT definition paper doesn’t address the problem of stemness. Are MSC stem cells or progenitors? This is still an ongoing discussion without official ISCT position. Clinical use of MSC has nothing to do with their “stem cell nature”. “Stemness” of MSC could be translated in multilineage differentiation potential. But we mostly use only one-lineage differentiation of MSC in clinic, such as osteogenic/ chondrogenic or don’t use their differentiation potential at all (see “medicinal cells“). Therefore, “multipotency” and “stemness” of MSC doesn’t reflect clinical application. Many academic researchers consider MSC as “progenitors” but not “stem cells”. Some industry developers also call MSC as “mesenchymal progenitor cells” (see Mesoblast).

I’d propose a discussion on this matter with release of ISCT position.

4. Fresh versus cultured MSC
ISCT definition doesn’t address the difference between fresh (native/ freshly isolated) and cultured MSC. The statement covers only MSC expanded ex vivo by defining “plastic adherence” as minimal criteria. This position ignores the fact that properties of native MSC, which reside in tissues, could be completely different from cultured MSC. For example, the recent discussion about CD34 as MSC marker, brought up this issue and highlighted the lack of this information in ISCT definition. Taking in account increasing interest to clinical application of freshly isolated MSC, this issue should be discussed.

I’d propose to address a difference between native and cultured MSC in revised version.

5. Markers
MSC defined by ISCT as positive for CD90, CD105 and CD73. In the light of recent research, some other markers could be considered. For example, MSCA-1, CD271, CD146 and some others. The difference in markers expression should be defined for fresh versus cultured MSC and, also, for MSC isolated from different tissues.

6. Heterogeneity
ISCT definition doesn’t address the difference between MSC, isolated from different tissues. The difference could be in markers expression, in differentiation potential and even in naming. For example, if cord blood doesn’t have a stroma, why should it be called MSC? Yet another question is how to define the cells expressing similar markers with MSC (ex: pericytes). I think the questions of heterogeneity should be addressed in the revision.

This is a pilot draft of proposal. Please discuss with me the necessity and content of revised criteria for MSC.

{ 2 comments… read them below or add one }

Corey Orava January 28, 2013 at 10:10 am

Great topic – surprised to see no comments so far. Below are my comments, point by point.
1. the poll is not for the “definition” of MSCs but rather the nomenclature… thus see 2.
2. I agree with the author that the use of “multipotent” is not necesary. In fact, I consider it misleading at best (in vivo, the differentiation of MSCs is NOT their major mechanism of action) and inaccurate at worst (many studies have provided evidence that MSCs, from a variety of tissue sources, are in fact pluripotent – yes, I know this is a contentious topic).
3.I agree, but I like the previous points, this deals mostly with semantics and does not necessarily reflect the clinical utility of the cells, as such, I would not consider this point a priority.
4. For me, the topic of fresh vs. cultured cells should be considered a priority. I would also go further and state that we need to spend more attention as to the CLINICAL/PHARMACOLOGICAL differences between cell populations after various culture regimes (e.g. few vs. many passages, serum vs. serum-free media etc).
5. While I think this is a topic that needs further study, there are a couple of reasons why I think we could be chasing our tails. First, many CD markers are inducible, as such using them to define a cell type will only be useful under specific conditions. Second, we do not know the function of many of these markers.
6. I wholeheartedly agree with the points made and would go a step further. Stem cells from any given tissue represent a heterogenous population. I consider it very naive for anyone to think that there is a single “best” cell type. No organ is made up of a single type of cell – why would we then think a single type of cell is best at healing/regenerating that tissue? I realize this will add significant complexity (studying populations of cells rather than individual cell lines) but I believe the clinical payoff(s) will be worth the effort.

Bottom line, organizations like the ISCT should be the vanguard of this technology, but resources should be prioritized to clinically-relevant topics rather than semantics.


Mark Weiss August 22, 2013 at 5:56 pm

I agree that further discussion is needed. Consensus would be nice, but realistically, forget about it and just be as clear and consistent as you can be. Thank you of leading us into the conversation with your comments.

My input.
Regarding your points 1 and 2 nomenclature. I like “multipotent mesenchymal stromal cells” for the time being. I think the adjective “multipotent” is important.

I agree that naive MSCs should be on the list, since these would be the parent cells derived from the donor. I think a problem here is that we are working with cultured MSCs, which are a contrivance or an artefact produced by in vitro manipulation and expansion. We can define naive MSCs, but does that category have value since it is a theoretical construct that evaporates as we try to capture it or might be changed by our attempts to examine it?

I argue that cultured MSCs, like ESCs, and iPSCs are artificial constructs of culture. A difference of MSCs and ESCs /iPSCs is that ESCs/ iPSCs are immortalized cells and more obviously artificial, and have a more highly defined specification and culture requirements.

3. Stemness. You got a couple of issues here. In vitro: You got a mixed population of multipotent, unipotent, null potent cells, aka MSCs in culture exposed to differentiation factors for weeks followed by non-quantitative readouts. Problems with interpreting the data: do we split confluent cultures, what about feeding the cultures, is it an issue that there are changes in cell number over time or that we are examining non-clonal populations. We can’t be sure that one cell truly has trilineage potential. In vitro assays could be run on replicate clonal populations,tightened up and made quantitative. In VIVO assays: We only defer to in vitro assays when in vivo assays fail us. Note that In vivo assays are the only definitive way to prove that a ESC or iPSC is pluripotent and the most rigorous assay requires that ESC or iPSC make virtually an entire mouse via tetraploid complementation or morula injection. For human pluripotent cells we can only perform teratoma assays which is the best data, but is a pale comparison to the mouse. IN VIVO stemness assays might be the answer. For expanded MSCs, Caplan has his in vitro osteogenic differentiation assay. True, it tests only osteogenesis, but it can be quantitative. I don’t know of any but I would suggest that quantitative in vivo assays for cartilage and fat lineages should be similarly derived until a quantitative assay that shows trilineage differentiation in vivo is available. Again the problem of purified input populations exists.

4. From my desire to move from preclinical into clinical trials, I applied to PACT and had a clinical translation advisory team assigned to evaluate my MSC data package. They informed me that I needed to repeat my preclinical data package using banked, cryopreserved MSCs instead of fresh, unthawed MSCs. We have experience with frozen/thawed MSCs and freshly expanded (never thawed cells) and strategically chose to work with freshly expanded cells. Of course, the rationale they gave me for using banked cells was pretty good, and they are PACT: MSCs can be expanded, cryobanked and aliquots removed for characterization/ validation of release criteria. Based upon this input, I suggest that cryopreserved MSCs be considered as additional category.

5. Markers. I agree that at CD271 and CD146 should be considered to be added to the list. I don’t know whether an input population of sorted CD271 and CD 146 cells is “pure” CD271 and CD146 after rounds of passage. It would be instructive to compare naive, expanded and expanded/ cryopreserved/ thawed for marker expression.

6. Heterogeneity. Its pretty clear that different culture conditions, isolation methods and donors produce different populations of cultured MSCs. It is no great leap to assume that MSCs are affected by tissue environments, cell culture substrates and the oxygen tension. MSCs are different from different age donors, too.


Leave a Comment

Previous post:

Next post: