Review Intravascular Mesenchymal Stromal/Stem Cell Therapy ... · (BIH), Berlin, Germany...

Review

Intravascular Mesenchymal Stromal/StemCell Therapy Product Diversification:Time for New Clinical Guidelines

Guido Moll ,1,2,3,* James A. Ankrum,4,5 Julian Kamhieh-Milz,6 Karen Bieback,7,8 Olle Ringdén,9

Hans-Dieter Volk,1,10,11,13 Sven Geissler,1,12,13 and Petra Reinke1,3,11,13

HighlightsMSC products have largely diversi-fied during the past decade, makingextrapolation about safety and effi-cacy from first-generation productsinappropriate.

MSCs and other blood non-residentcellular therapeutics display differentdegrees of incompatibility with humanblood, which compromises their safetyand efficacy.

Intravascular infusion is the most popular route for therapeutic multipotentmesenchymal stromal/stem cell (MSC) delivery in hundreds of clinical trials.Meta-analysis has demonstrated that bone marrow MSC infusion is safe. It isnot clear if this also applies to diverse new cell products derived from othersources, such as adipose and perinatal tissues. Different MSC products displayvarying levels of highly procoagulant tissue factor (TF) and may adverselytrigger the instant blood-mediated inflammatory reaction (IBMIR). Suitablestrategies for assessing and controlling hemocompatibility and optimized celldelivery are crucial for the development of safer and more effective MSCtherapies.

TF is the major determinant of cell pro-duct hemocompatibility, and thisshould be routinely monitored in alltherapeutics intended for intravasculardelivery.

MSC products from different tissuesources display high variability in TFexpression, with potential lethal con-sequences for patients when infusedsystemically.

Once aware of the problem, a largearray of product and process innova-tions became available to improve theclinical delivery of systemically infusedcellular therapeutics.

1Berlin-Brandenburg Center forRegenerative Therapies (BCRT),Charité Universitätsmedizin Berlin,corporate member of Freie UniversitätBerlin (FUB), Humboldt-Universität zuBerlin (HUB), and Berlin Institute ofHealth (BIH), Berlin, Germany2Berlin-Brandenburg School forRegenerative Therapies (BSRT),Charité Universitätsmedizin Berlin,corporate member of Freie UniversitätBerlin (FUB), Humboldt-Universität zuBerlin (HUB), and Berlin Institute of

Risk of Severe Adverse Events with Intravascular MSC TherapeuticsBone marrow multipotent mesenchymal stromal/stem cells (BM-MSCs) are one of the mostpromising cellular therapies (see Glossary), although clinical implementation remains chal-lenging [1]. Recently, a growing number of novel MSC products derived from tissues other thanBM, such as adipose tissue (AT) and perinatal tissue (PT), have entered clinical investigation,now accounting for �50% of products applied in clinical trials (Figure 1A) [2]. Intravascularapplication through intravenous or intra-arterial infusion is the most popular route of delivery fordiverse MSC products and has resulted in mixed clinical outcomes (Figure 1B) [3–7].

Meta-analysis has shown that infusion of first-generation BM-MSC therapies is safe [8], butuncertainty exists regarding (i) hemocompatibility and optimal delivery of MSC products, (ii)adverse events after infusion of large cell doses, and in particular (iii) the safety and efficacy ofnovel AT- and PT-derived MSC products [3,5,9–13]. Potentially lethal adverse events, suchas thrombosis and embolization, can result from MSC incompatibility with the innate immunecascade systems of the blood, and have been repeatedly reported in both animals andhuman patients [7,10–25]. These reports of adverse events and single incidents of fatalhuman cases potentially caused by thrombotic complications/embolism with diverse newMSC products highlight the clinical need for new standards in hemocompatibilityassessment.

To ensure patient safety, Cox et al. recently emphasized that all cellular therapies intendedfor intravascular delivery should be subject to hemocompatibility screening before clinicalapplication [13]. Product and process optimization and new clinical guidelines will benecessary to supplement the current minimal standards for MSC product characterization(see Clinicians Corner) [1,26,27].

Trends in Molecular Medicine, February 2019, Vol. 25, No. 2 https://doi.org/10.1016/j.molmed.2018.12.006 149© 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

Health (BIH), Berlin, Germany3Department of Nephrology andInternal Intensive Care Medicine,Charité Universitätsmedizin Berlin,corporate member of Freie UniversitätBerlin (FUB), Humboldt-Universität zuBerlin (HUB), and Berlin Institute ofHealth (BIH), Berlin, Germany4Roy J. Carver Department ofBiomedical Engineering, University ofIowa, Iowa City, IA, USA5Fraternal Order of Eagles DiabetesResearch Center, PappajohnBiomedical Institute, University ofIowa, Iowa City, IA, USA6Department of Transfusion Medicine,Charité Universitätsmedizin Berlin,corporate member of Freie UniversitätBerlin (FUB), Humboldt-Universität zuBerlin (HUB), and Berlin Institute ofHealth (BIH), Berlin, Germany7Institute of Transfusion Medicine andImmunology, Medical FacultyMannheim, Heidelberg University,Mannheim, Germany8German Red Cross Blood DonorService Baden-Württemberg-Hessen,Mannheim, Germany9Translational Cell Therapy Research(TCR), Department of Clinical Science,Intervention and Technology(CLINTEC), Karolinska Institutet,Stockholm, Sweden10Institute of Medical Immunology,Charité Universitätsmedizin Berlin,corporate member of Freie UniversitätBerlin (FUB), Humboldt-Universität zuBerlin (HUB), and Berlin Institute ofHealth (BIH), Berlin, Germany11Berlin Center for AdvancedTherapies (BECAT), CharitéUniversitätsmedizin Berlin, corporatemember of Freie Universität Berlin(FUB), Humboldt-Universität zu Berlin(HUB), and Berlin Institute of Health(BIH), Berlin, Germany12Julius Wolff Institute (JWI), CharitéUniversitätsmedizin Berlin, corporatemember of Freie Universität Berlin(FUB), Humboldt-Universität zu Berlin(HUB), and Berlin Institute of Health(BIH), Berlin, Germany13Equal contribution senior authorship

*Correspondence:[email protected] (G. Moll).

Clinical MSC Products Have Greatly Diversified in the Past DecadeA great diversification in production processes and tissue sources for the isolation of MSC-therapeutics has taken place in the past decade. According to the FDA [2] and an updatedsummary (Figure 1A), until 2008 MSC-like investigational new drugs almost exclusively com-prised BM-MSCs. This preference is rooted in the history of MSCs that were originallydescribed by Friedenstein as BM-resident cells supportive of hematopoiesis [5,28], thusmaking BM the first choice for therapeutic cells.

Initial studies focused on the multipotent differentiation potential of BM-MSCs to give rise toskeletal tissues [29]. However, results from multiple groups demonstrated that even minimallyexpanded BM-MSCs show little long-term engraftment and tissue-forming capacity uponintravascular infusion [30–32], which is the most common route of cell administration(Figure 1B). With long-term engraftment being limited, scientific focus redirected to Frieden-stein’s second discovery: the trophic activity of BM-MSCs [28]. This rationale aims to employthe beneficial trophic and immunomodulatory activity of MSCs for clinical use in regenerativeapproaches and the treatment of severe immune ailments, and these presently account for alarge share of ongoing clinical trial activity [6].

Another crucial paradigm shift was the intriguing conception of MSC in vivo identity as peri-vascular pericytes [33,34]. Indeed, the discovery that perivascular cells meeting the MSC minimalcriteria set out by the International Society for Cellular Therapy (ISCT) [26] could be isolated frommany vascularized tissues, and in high abundance in the case of AT and PT tissues, greatlydiversified the study and application of MSC products. The highly vascularized PTs in particularharbor large quantities of pristine MSC-like cells with distinct phenotypes [35–38]. Several potentcell types, such as placenta-derived decidual stroma cells (DSCs) [39], can easily be sourced andminimally expanded for clinical use at early passage, and this has been suggested to yield betterefficacy in some clinical applications [40].

Since expansion of the MSC field from only BM to include a multitude of MSC products fromdiverse sources, similarities as well as phenotypic and functional differences have been identified[20,35,36]. Although MSC products are minimally characterized by surface marker expressionand differentiation capacity [26], their apparent differences require meticulous individual reas-sessment of therapeutic safety and efficacy for different routes of delivery because equivalencecannot be assumed. Crucially, intravascular infusion has remained very popular, even though themajority of MSC products tested today are no longer derived from BM (Figure 1B).

Although diversification beyond BM overcomes a number of problems (e.g., invasive harvestingtechnique, limited starting material, and potency issues), equivalence in hemocompatibility andsafety cannot be assumed for MSCs from alternative tissue sources expressing higher levels oftissue factor (TF/CD142) [13,20–22,41–43], a key trigger of coagulation [41]. It is not clearwhether past evidence of BM-MSC safety can be extrapolated to products from other sources,and how this interlinks with their efficacy [1]. In islet cell transplantation, the triggering of adverseinnate immune responses, termed the instant blood-mediated inflammatory reaction(IBMIR) [44], has been identified as a crucial threat to graft survival and function, thus beingidentified as an important target for therapeutic cell product optimization [44–48].

It could be argued that the unsatisfactory in vivo engraftment, compromised bioactivity, and thequestionable safety profile of many non-hematopoietic extravascular (stem) cell therapeuticsbeing delivered by intravascular infusion presents an undesirable obstacle to their clinicalapplication [1,3–6,48,49].

150 Trends in Molecular Medicine, February 2019, Vol. 25, No. 2

GlossaryCellular therapy: a therapy (e.g.,stem cell therapy or immunotherapy)which employs cellular material (e.g.,MSCs, immune, or islet cells) that isdelivered into a patient by differentroutes of application (e.g., systemicintravascular injection or localtransplantation) in an attempt to cureor alleviate illness.Coagulation: the process by whichblood changes from a liquid to a gel,by forming a blood clot, for exampleupon blood vessel injury, promotinghemostasis and cessation of bloodloss from a damaged vessel. Underpathological conditions, clotting maylead to thrombosis, the cessation ofblood flow inside blood vessels.Hemocompatibility: medicaldevices that contact blood duringclinical use must be hemocompatible(biocompatible with blood; ISO10993-1/4) to not adversely interact withblood components, and this may alsoapply to therapeutic cells.Instant blood-mediatedinflammatory reaction (IBMIR): adetrimental instant innate immuneattack that can compromise thesafety and efficacy of infused cellulartherapeutics and medical devices incontact with blood.Perinatal tissue: the perinatalperiod commences at 22 completedweeks (154 days) of gestation andends 7 completed days after birth,with the term perinatal tissue ‘stem’

cells referring to progenitor cellsobtained from developmentally younggestational tissue (e.g., placenta,decidua basalis, chronic villi, amnion,and amniotic fluid, as well asumbilical cord matrix and cord blood)obtained around the time of birtheither by cesarean or natural birth.Tissue factor (TF/CD142/coagulation factor III): aglycoprotein that triggers the extrinsicpathway of the coagulation cascadeby initiating thrombin formation fromthe zymogen prothrombin [41].Expressed in cells that are normallynot exposed to flowing blood, suchas subendothelial cells (e.g., smoothmuscle cells) and cells surroundingthe blood vessels (e.g., fibroblasts),while the inside of blood vessels islined with endothelial cells that donot express TF unless exposed toinflammatory mediators such asTNF-a.

Intravascular MSC Therapeutics Display Incompatibility with Human BloodIt has been a matter of debate whether MSCs are suitable for contact with blood in vivo, andwhether this changes upon expansion ex vivo (Figure 2) [50–52]. MSCs from different tissuesources differ largely in their hemocompatibility [7,13,20–22], and infusion of procoagulant TF-expressing cell products, such as MSCs, hepatocytes, and pancreatic islets, triggers innateimmune attack (IBMIR) that compromises cell engraftment, safety, and efficacy (Figure 3;Tables S1 and S2 in the supplemental information online) [44,53,54].

Shapiro et al. first reported causality for clinical cells’ triggering of IBMIR, their compromisedbioactivity (e.g. Loss of insulin production) and subsequent loss in clinical efficacy (e.g. Reversalof insulin independence), thus requiring infusion of multiple cell grafts [45–47]. Similarly, infusedMSCs are rapidly embolized and destroyed in the microvasculature, at least partly caused bytheir triggering of IBMIR, which compromises their bioactivity and may lead to thromboembo-lism when used without anticoagulants [10–12,14–16,55,56].

Intrinsic Hemocompatibility and Incompatibility of Stem Cell Types In VivoSeveral progenitor cell types with (potential) circulating phenotype are of interest when dis-cussing the hemocompatibility aspects of cellular therapies (Figure 2A): (i) hematopoietic stemcells (HSCs), (ii) endothelial progenitor cells (EPCs), and (iii) various types of MSCs that areresident throughout different organs and tissue locations within the organism [57–59]. Acommon feature of progenitor cells is their intrinsic tropism and migration towards woundsand inflammatory sites [49]. To reach their target, progenitor cells can be selectively mobilizedfrom their in vivo reservoirs and attracted by specific cytokine and chemokine gradientsproduced by the injured tissue and the local immune infiltrate [59].

Cellular homing responses are enhanced by blood activation products that are released uponvessel rupture [60]. MSCs express receptors for complement activation products (e.g., C3a,C5a) that partly guide their chemotactic responses [61] to harness their tissue repair propertiesand safeguard vessel integrity [7,62]. A major difference between HSCs/EPCs and MSCs,however, is that the former two are well known to mobilize into the bloodstream from their invivo reservoir and circulate in circadian fashion [59,63,64], while this is less clear for MSCs[1,50–52]. MSCs are predominantly extravascular cells that reside in the perivascular niche, andare commonly not found in, or in direct contact with, the blood (Figure 2A) [5,65].

The circulatory phenotype of MSCs is hotly debated, but is not yet a broadly accepted feature ofthese cells [50–52]. Although growing evidence suggests that native MSCs can at least betransiently found in blood, for example following severe fractures [65,66], only a small portion ofclinical MSCs survive intravascular infusion and extravasate out of the vasculature [16,67]. Themajority of the infused therapeutic MSCs are lost due to triggering of innate and adaptiveimmune cascades, embolization, and micro-ischemia [16]. By contrast, endothelium andHSCs/EPCs exhibit high hemocompatibility because they express blood regulatory moleculesthat keep innate immune cascades in check (Figure 2B). The vascular endothelium, located atthe interface between blood and tissue, is indeed the best-studied model system for under-standing the molecular mechanisms that form the hemocompatible interface [41,68,69].

In conclusion, the in vivo phenotype of tissue-resident MSCs is still poorly defined regardingtheir hemocompatibility. Molecular profiling has revealed that culture-expanded MSCs partlylack blood regulatory molecules and express higher levels of extravascular procoagulantfactors compared to the prototypic hemocompatible endothelial cells (ECs) (Figures 2B and3; Table S1).

Trends in Molecular Medicine, February 2019, Vol. 25, No. 2 151

Product diversifica on Transla onal pipeline Product delivery Route of MSC administra on

Clinical choice of MSC products1994–2009 2009–2013 2013–2018

Donor AT

PT

BM Preclinicalproof of concept

Safety andpotency tes ng

Regulatoryapproval

Cell manufractureand clinical trials

Local

Intravascular

Pa ent

ND/otherAdipose ssuePerinatal ssueBone marrow

Mile

ston

es to

war

dsim

prov

ed d

eliv

ery

ofcl

inic

al M

SC p

rodu

cts

Cum

ula

ve c

linic

al tr

ials

(ong

oing

and

com

plet

ed)

Trial start year

800

600

400

200

0

1994

–200

420

0520

0620

0720

0820

0920

1020

1120

1220

1320

1420

1520

1620

1720

18

Intravascular delivery (48%)Local delivery (49%)ND/other (3%)

38%56%57%

27%27%1%

22%16%3%

13%1%39%

(A) (B)

Updates on improvedMSC characteriza on

First report on human MSCincompa bility with blood

Product diversifica on: first humancase reports of severe side effects

Consensus paper on MSC minimal criteria.Alarming animal studies on MSC emboliza on

First reports on intravascular infusion of BM-MSCsdemonstra ng clinical feasiblity of the approach

Figure 1. Increasing Product Diversification in Clinical Use of Mesenchymal Stromal/Stem Cell (MSCs). (A) Therapeutic MSCs were initially isolated frombone marrow (BM) [29]. With growing interested in their clinical use and the discovery of MSCs in other vascularized tissues, an increasing contribution from adipose andperinatal tissue (AT and PT) MSCs became evident, starting in 2009. ND, not disclosed. (B) The number of trials employing MSC products has increased steadily, withcells being administered both locally and intravascularly (intravenous or intra-arterial infusion) with an approximately equal distribution of the two modes of therapeuticdelivery. Methods: clinical trial data were collected from the clinicaltrials.gov registry on June 15, 2018. Trials were selected for analysis by searching for [‘MesenchymalStem Cell’ OR ‘Mesenchymal Stromal Cell’ OR ‘Multipotent Stromal Cell’ OR ‘Bone Marrow Stromal Cells’]. The search returned 840 results. The clinical trial descriptionand intervention used in each trial were analyzed to identify the source of the MSCs and the route of administration; 52 trials were found to not use MSCs as anintervention and were excluded from the analysis, leaving 788 trials for analysis.

Importantly, their intrinsic expression profiles depend on the tissue source or starting materialthe MSC products were derived from, making them more or less vulnerable from a hemo-compatibility perspective [7,13,20,62,70].

Tissue Factor in Clinical MSC Products Is the Major Trigger of IBMIRThe triggering of IBMIR by MSC products and other cellular therapies is correlated with theirexpression of TF (Figure 3A and Table S1), a highly potent trigger of the TF pathway of coagulation[41,44,53]. Importantly, both TF expression and triggering of IBMIR by MSC products depend onthe tissue source the cells were derived from. Compared to BM-MSCs, AT- and PT-derivedproducts demonstrate higher TF expression and reduced hemocompatibility, with substantialdonor variation and culture passage-dependent TF induction [7,13,18,22–24].

Different approaches to block the TF pathway [e.g., TF blocking antibody, inactivated/activatedcoagulation factor VII (FVIIai), and FVII-deficient plasma] demonstrated that a large share of theprocoagulant activity of various clinical MSC products could be attributed to triggering theextrinsic TF pathway of coagulation [7,13,20,23].

Numerous stimuli, for example proinflammatory cytokines such as interleukin-1 and tumornecrosis factor a, as well as blood activation products such as thrombin, can promote TF


Progenitor reservoirs (e.g. BM & PT) Differen�ally expressedprothrombo�c factors

Differen�ally expressedan�thrombo�c factors

TM

MSC

EC

TM

PG12

iNOS

eNOS

PG12

CD46CD55

CD46CD55

TFPI

TFPI

TF

TF

Col-1

Col-1

Basallamina

Intravascular ECs possess lower prothrombo�c andhigher an�thrombo�c proper�es than perivascular MSCs

Intravascular and extravascularprogenitor cell recruitment

Micro-vascular

EC

Intravascularcircula�ngHSC/EPC

PerivascularMSC

(A) (B)

Figure 2. Progenitor Recruitment and Hemocompatible Biological Interface. (A) Human blood containscirculating progenitor cells (e.g., HSCs and EPCs) that are released into the blood from their reservoirs, for examplebone marrow (BM) and perinatal tissue (PT). It is not yet clear if MSCs circulate within the blood or if they migrate though theperivascular compartment. (B) The vascular endothelium is the ‘gold-standard’ of thromboregulation and serves as ahemocompatible interface between blood and tissue that prevents innate immune cascade activation [69]. IntravascularECs exhibiting low prothrombotic and high antithrombotic properties. By contrast, MSCs express higher levels ofprothrombotic tissue factor (TF/CD142) and collagen type-1 (COL-1), two key triggers of the extrinsic tissue factorand intrinsic contact activation pathways of coagulation, respectively. TF levels greatly vary between MSC sources.Abbreviations: CD46, membrane cofactor protein; CD55, decay accelerating factor; e/iNOS, endothelial or inducible nitricoxide synthase; EC, endothelial cell; EPC, endothelial progenitor cell; HSC, hematopoietic stem cell; MSC, mesenchymalstromal/stem cell; PGI2, prostacyclin or prostaglandin I2; PTGIS, prostaglandin inducible synthase; TFPI, tissue factorpathway inhibitor; TM, thrombomodulin.

expression through binding to their respective receptors (Figure 3B) [41,71]. This triggersintracellular signaling cascades leading to TF (coagulation factor III, F3) promoter activation,mRNA production, and intracellular TF production. Importantly, TF can be found (i) stored incytoplasmic intracellular vesicles, (ii) integrated into the cell membrane, or (iii) exported into theextracellular compartment as soluble TF or incorporated within extracellular vesicles (EVs) [71].

TF is a potent trigger of the extrinsic TF pathway of coagulation, leading to thrombin generation,platelet activation, and fibrin crosslinking. In addition, stromal cells express varying degrees ofstrongly negatively charged stromal extracellular matrix (ECM) components, such as variouscollagens, which can trigger the intrinsic contact-activation pathway of coagulation [7]. Trig-gering of coagulation is mediated via activated intermediates [e.g., activated coagulation factorVII (FVIIa), FXIIa, FXIa, FVIIIa, FXa, and FVa] and is regulated in a highly complex manner [72].

Another major arm of IBMIR is the complement cascade, which is initiated via triggering of theclassical, alternative, and lectin pathways [70,72–74]. These pathways converge at the com-plement factor 3 (C3) convertase, converting complement factor C3 into its activation productsC3a and C3b. The larger fragment, C3b, is a key element in forming the C5 convertase,converting C5 into C5a and C5b, the initiator of the membrane attack complex (MAC) via


IntravascularMSC delivery

Binding of clothing factorsto therapeu�c cells

IBMIR

MSCengra�ment

Tissue factor (%)Passage number25

2525

0P1-3 P4-6 75

7575

50

5050

100

100100

BMPTAT

Resp

ec�

vesi

gnal

inte

nsit

y (%

)

TF e

xpre

ssio

n[%

]

MSC heterogeneityin TF expression

Tissue factor(TF)

BM

PT

AT

IBMIR-targetedcompromised MSC

Innate and adap�veeffector cell modula�on

B cellT cell

Dentri�c cells

Macrophage

Monocyte

NeutrophilMast cell

IBMIR

External s�muli triggerTF expression, e.g.

IL-1

IL-1R

TNF-α Thrombin EV

PAR1TNFR1

Intracellular signaling

TF promoter

Nucleus

Cytosol

Intracellular TF pool

TFPoly-A

Poly-APoly-A

Extracellular TF

Surface TF

Soluble TF

(B)

(A)

Figure 3. Tissue Factor (TF) Expression in Mesenchymal Stromal/Stem Cell (MSC) Products and Triggeringof the Instant Blood-Mediated Inflammatory Reaction (IBMIR). (A) MSC products isolated from bone marrow (BM),perinatal tissue (PT), or adipose tissue (AT) express varying degrees of tissue factor (TF), a key trigger of the clottingcascade, which can be modulated by the degree of cell expansion in vitro (overviewed in Table S1). MSC TF expression ispositively correlated with their triggering of the IBMIR, which is inversely correlated with the short-term engraftment of thecells. The triggering of IBMIR can have diverse effects in vivo, such as modulating innate and adaptive effector cells. (B) TFsignaling can be activated by numerous stimuli through binding of specific receptors, inducing intracellular signaling, TF(F3) promoter activation, gene transcription, mRNA production, and protein expression [41]. TF can be intracellular (eithercytoplasmic or within multivesicular bodies, MVBs), on cell surfaces (within the cell membrane or as encrypted TF), andextracellular (released as alternatively spliced soluble TF, or within microparticles and extracellular vesicles, EVs) [71].Examples of TF stimuli and their corresponding receptors are: IL-1b/IL-1R, interleukin 1b and IL-1 receptor; TNF-a/TNF-R1, tumor necrosis factor a and TNF receptor 1; thrombin/PAR1, thrombin and protease-activated receptor 1.

incorporation of C6–9. The complement system is also regulated in a complex manner, forexample through the actions of CD35, CD46, CD55, CD59, FI, and FH.

Diverse Clinical MSC Products Trigger IBMIR upon Intravascular DeliveryEarly clinical trials with BM-MSC therapeutics demonstrated a good safety profile and no majorinfusional toxicity in diverse treatment indications [8]. This is in line with the low to absentexpression of TF on minimal expanded BM-MSCs (Table S1) and the common systemic


anticoagulation treatment of patients treated with first-generation MSC products in thereported trials (Table S2). It is not clear if this generally also applies to diverse new MSCderived from other tissue sources, or how an unfortunate combination of high doses of stronglyTF-expressing cells would fare in patient populations without anticoagulation (examples aregiven in Table S2).

Many of the acutely ill patients in trials that reported safety of MSCs in terms of infusion toxicity[e.g., acute myocardial infarction (AMI), acute lung injury (ALI), acute respiratory distresssyndrome (ARDS), peripheral artery disease (PAD), and ischemic stroke] were already antico-agulant-treated to some degree (e.g., anti-platelet drugs, low molecular weight heparin anddextran sulfate), and the research findings may not be easily generalizable to systemicallyactivated/proinflammatory patients who do not receive anticoagulation treatment at the time ofcell infusion (Table S2) [8,75–78].

Moll et al. first reported the triggering of IBMIR by BM- and PT-derived clinical MSC productsinfused into patients (Tables S1 and S2) [7,20], and other groups rapidly confirmed thesestudies, demonstrated a certain degree of incompatibility of culture-expanded MSCs withblood, which is modulated by multiple parameters including donor tissue type and ex vivoprocessing [17,19,22,23]. While minimally expanded BM-MSCs triggered weak IBMIRresponses in vitro and in vivo [7,13,17,20,22], PT-derived DSCs triggered stronger responseswhen tested without additives [20]. Nonetheless, low doses of DSCs triggered only weak IBMIRresponses in patients, and this may be attributed to optimized reconstitution and application(Tables S1 and S2) [39,79].

An increased rate of infusion reactions was reported for high doses of other PT-derivedproducts [77,78]. The clinical need for increased caution and optimized delivery procedureswith PT-MSCs is also illustrated by a recent report [12] documenting venous thrombosis in theforearm of patients treated with umbilical cord (UC)-MSCs. Both patients developed painfulforearm swelling at a site distant from the injection site, with Doppler ultrasound imagingconfirming intravenous clot formation and baseline coagulation tests documenting elevated D-dimer (fibrin degradation fragment) levels. Warfarin and urokinase were used as thrombolytictherapy for clot reduction. By contrast, other investigators did not observe adverse events uponUC-MSC infusion (Table S2) [80].

AT-MSCs were shown to exhibit highly procoagulant activity and even lethal effects uponinfusion in preclinical models (Table S1) [18,21,81]. Case reports of peripheral microthrom-bosis, pulmonary embolisms, and even suspected cases of death in patients receiving AT-MSCinfusions call for caution with intravascular delivery (Table S2) [10,11,82]. By contrast, theInternational Federation for Adipose Therapeutics and Science (IFATS) did not report anincreased risk for adverse events/infusion toxicity with AT-derived MSCs in clinical studies[83]. Interestingly, a recent dose-escalation study documented that intravenous infusions ofAT-MSCs from healthy donors are well tolerated in humans up to 4 � 106 cells/kg body weight,when a significant increase in clotting parameters and fibrinolytic response became apparent[25]. Importantly, AT-MSC hemocompatibility may also be influenced by donor comorbiditiessuch as diabetes [82,84,85].

Several studies addressed the impact of culture expansion, culture media, freeze-thawing,washing, and the cell suspension buffer for MSC triggering of IBMIR (Table S1) [23,40,86,87].Analysis of the supplements used for MSC expansion documented variable degrees of clottingin vitro [23]. A mild increase in adverse events was reported for infusion of large doses of


cryopreserved MSCs compared to fresh cells [88]. Freeze-thawed MSCs triggered a strongerIBMIR when washed with buffer containing human blood type AB plasma instead of humanserum albumin (HSA) [40,86]. The cell suspension medium impacts on the pulmonary toxicity ofMSC products, with low-dose heparin being highly beneficial, avoiding 95% of clinical symp-toms [87], and low molecular weight dextran sulfate (LMWDS) was effective in controlling IBMIRin hepatocyte transplantation [89]. We thus advocate regular use of low-dose anticoagulantssuch as heparin in the clinical setting, if applicable, and the reporting of anti-coagulant usewithin clinical trial protocols (Table S2) [20,89,90]

In conclusion, multiple reports have indicated a favorable outcome for minimally expandedfresh cells prepared with non-immunogenic supplements, and, if applicable, that either patientsor cell infusion buffers should be treated with anticoagulants, with the tissue source the MSCsare derived from being a major defining element.

Process and Product Innovations to Optimize Clinical Cell DeliverySeveral strategies and delivery methods can be used to monitor, antagonize, or avoid triggeringthe IBMIR (Box 1; Figures 4 and 5 ). Improvements in the large array of existing and newlydeveloped MSC products can be achieved through process optimization and advancedbioengineering (e.g., better hemocompatibility), which may not only result in better safetybut also improve efficacy [4,48,91,92]. According to ISO 10993-1/4, medical devices thatcontact blood in a clinical routine must be subjected to hemocompatibility testing (Box 1; WallinR.F. 1998: http://www.mddionline.com/practical-guide-iso-10993-4-hemocompatibility).

Box 1. How to Conduct Hemocompatibility Assessment of Cell Products?

According to ISO 10993-1/4, medical devices that contact blood in a clinical routine must be subjected to hemo-compatibility testing. Therapeutic MSC products can be tested with the following methods to obtain an adequate safetyestimate.

(i) Hemocompatibility Profiling and Ex Vivo Testing with Extracorporeal Systems

Hemocompatibility marker monitoring (e.g., detection of gene expression or surface markers) (Figure 4A) [26,27] andhemocompatibility assessment with assays employing human blood from healthy, drug-free, blood type AB donors(Figure 4B) [7,23], such as the Chandler whole blood loop [7,86], viscoelastic rotational plasma-thromboelastometry(ROTEM) and thromboelastography, could be added to the current standards for MSC characterization [13,17–20,23,24,81].

(ii) In Vivo Hemocompatibility Testing in Preclinical Animal Models

In vitro assays are often paralleled by in vivo models [18,19,21,23,24]. Large doses of BM-MSCs were found to triggerthrombosis [24], but heparin anticoagulation prevented clotting and adverse events, and increased the efficacy of thecells [19,23,24]. Detrimental effects often become apparent at high cell doses (Tables S2 and S3) [24,78,88]. Commonlyused doses of MSCs in human are 1–10 � 106 cells/kg of body weight, whereas typical doses tested in mice are 4–5-fold higher than in humans (e.g., 0.1–1.0 � 106 cells/animal, corresponding to 5–50 � 106 cells/kg for a mouseweighing 20 g).

(iii) Testing of Complement Compatibility In Vitro and In Vivo

The complement compatibility of MSCs can be tested [7,40,62,70,86,97–99], but MSC testing for example in Chandlerblood loops only allows C3a/C5a detection, whereas MSC coculture with lepirudin-anticoagulated blood allowsdetection of C3a/C5a and CD11b priming on peripheral blood mononuclear cells (PBMCs) [62], and MSC exposureto complement-active serum allows detection of C3a/C5a and C3 fragments (e.g., C3b, iC3b, and C3d) withquantification of cell number and cell viability [7,40,62,70,86,97,98]. The complement susceptibility of MSCs andinterventional strategies were also validated in vivo to confirm prior in vitro findings [70,98,99].


(A)

(B)

Expression profiling of diverse MSC products

Hemocompa�bility assessment in vitro and in vivo

Therapeu�c cell gra�

Primary cells Cells in culture Expression profiling

Mat

rix

of M

SC p

rodu

ctpa

ram

eter

s of

inte

rest

Clinicalsafety

threshold

An�coagula�onRecommended Mandatory

Intensity of IBMIR in vitro IBMIR-related outcome in vivoTime (days)25 50 75 100

Ani

mal

sur

viva

l (%

) 100

75

50

25

5 10 15 20 25

With heparin

w/o heparin

Deple�on ofTF+ MSCs

In vivotes�ngChandler

loop

ROTEM

Figure 4.

(Figure legend continued on the bottom of the next page.)

Standardized Hemocompatibility Testing of Clinical Mesenchymal Stromal/Stem Cell (MSC)Products. (A) Expression profiling of hemocompatibility markers can be on conducted on diverse MSC products (e.g., primary cells directly after isolation from tissue, culture-expanded cells at different passages, and final product in itstherapeutic formulation). (B) The interaction of MSC products with (human) blood can be studied with in vitro test systems[e.g., viscoelastic rotational plasma-thromboelastometry (ROTEM) and Chandler blood loop] or animal models [48,53].These in vitro systems allow testing of a complex assay matrix with multiple test conditions, for example, clinically relevantdoses of MSCs sourced from different tissues (bone marrow, BM; perinatal tissue, PT; adipose tissue, PT, etc.), tested atlow or high passage (LP or HP), for fresh or cryostorage-derived thawed cells (F or T). The given examples of intensities for


Clinical Cell Reconstitution, Anticoagulants, and Therapeutic OutcomeTargeted optimization and standardization of cell reconstitution, formulation, and deliveryare key aspects in steering MSC product safety and efficacy [92]. A highly confoundingfactor in clinical use of MSCs is the lack of information regarding reconstitution buffers usedfor cell injection or transfusion, with great variation in procedures between clinical centers.For in vitro testing, freshly harvested cells are most commonly used, and these areresuspended in sterile buffers such as phosphate-buffered saline (PBS), which does notreflect the clinical reality. By contrast, clinically applied cells are commonly derived fromcryostorage, either with or without washing to remove cell debris and residual cryopro-tectants such as dimethyl sulfoxide [40].

The clinical cell washing/infusion buffer is often supplemented with additives such as proteinsources (e.g., blood type AB plasma or HSA) and anticoagulants (e.g., heparin or EDTA)(Table S2) to optimize the stability, tolerability, and performance of the cell graft, which shouldalso be evaluated during in vitro testing [40,86,87]. Several reports have emphasized thebenefits of replacing AB plasma (containing ABO antigen, complement, and clotting factors)with ABO-neutral HSA and adding low-dose anticoagulants to MSC therapies[17,19,20,39,79,86]. These additives block complement and coagulation activation, andminimize the risk of adverse IBMIR-triggering by TF-bearing cell grafts [20,48]. Interestingly,Stephenne et al. first suggested bivalirudin in conjunction with heparin to suppress theprocoagulant activity of non-hematopoietic MSC-like cells [17].

Analysis of the safety and efficacy of PT-derived DSCs has shown good safety and firsthints of better efficacy compared to conventional treatment [39,79,93]. All patientsresponded to DSCs at 1–2 � 106 cells/kg and had a 1 year survival comparable to controlswithout graft-versus-host disease (GvHD), particularly when the cells were supplementedwith buffer containing HSA and low-dose heparin [39]. Patients receiving DSCs preparedwith HSA had a 1 year survival of 76%, and 100% responded (21/21 patients) compared toa 1 year survival of 47% and 59% responding (10/17 patients) to DSCs prepared with ABplasma, respectively [39]. In steroid-resistant GvHD patients, DSCs suspended with HSAor AB plasma were superior to BM-MSCs or steroid-resistant controls without celltreatment (73%, 31%, 20%, and 3% 1 year survival, respectively). Thus, avoiding ABplasma and tailored supplementation with anticoagulants may be a simple way to improvethe efficacy of MSC therapeutics [86]. To our knowledge, no other MSC product hasachieved this efficacy so far and Phase III studies to confirm these findings are urgentlyneeded.

Novel Strategies for Cell Modification and Alternative Routes of DeliveryGalipeau and Sensébe recently gave a perspective on the clinical challenges and therapeuticopportunities in MSC therapy, addressing MSC dosing, fitness, potency assays, mechanism ofaction, and successful design of clinical MSC products and trials, giving a new and constructiveimpetus for moving the field forward [1]. They highlighted the concept of ‘Plazentallen as theRosetta Stone of MSC function’.

triggering of the instant blood-mediated inflammatory reaction (IBMIR) in vitro are estimates based on values reported inpublished literature (Table S1), depending, among others, on the assay system, cell type, passage number, and cellpreparation. This procedure may be informative for establishing safety margins for systemic cell application (e.g.,recommended or mandatory anticoagulation) or to facilitate decisions on alternative approaches, such as testing ofsuitable interventions in preclinical animal models (e.g., therapeutic cells with or without heparin, or depletion of tissuefactor (TF)-positive cells within the cell graft).


Key Figure

Mesenchymal Stromal/Stem Cell (MSC) Product and Process Innova-tions for Optimized Cell Delivery

(A)

(D) (C)

(B)The pa ent

The solu on The problem

The productAlterna ve routes of product delivery Safety of diverse MSC products

Tissue factor(CD142)

Blood incompa bility !

IBMIR

Tissuefactor

(CD142)

Intravascularinfusion

Intrathecal orintraperitoneal Extracorporeal

dialysis systems

Local injec on or implanta onof cells or cell clusters

Product innova onaTF

mAb iaFVII An coagulatants(e.g., heparin)

Ca2+ chelator(e.g., EDTA)

Membranemodifica on

Minimalexpansion

asRNA

Monitor IBMIRin vitro and in vivo

CRISPR/Cas9genome edi ng

TF+/- cellselec on

AbolishTF ac ons

SuppressTF ac ons

Bone marrow

Perinatal sssue

Adipose ssue

Complement andcoagula onac va on

Crosslinkedfibrin

Cell damage and emboliza on in the microvasculature

ROTEM

Mousemodels

Chandlerloop

Figure 5. (A)Dependingonthetarget indication, intendedsite,andmodeofaction,alternativeroutesandmodesof intravascularornon-intravascular celldelivery can be applied. These entail alternative routes (e.g., intratrahecal and intraperitoneal); local site deliveryvia intramuscular injection, implantation, or biomaterial assisted-delivery of MSCs (e.g., cells contained in fibrin glue or seeded ondecellularized extracellular matrix patches); and via extracorporeal dialysis for detoxification, where cells are seeded into hollow-fiberdialysis cartridges, the cell-containing bioreactor module, which is used to supplement conventional dialysis systems [94,95]. (B)Different levels of tissue factor (TF/CD142) expression in diverse types of MSC products, for example sourced from bone marrow,adipose tissue, or perinatal tissues, and (C) their consequent predominantly TF-mediated triggering of the instant blood-mediatedinflammatory reaction (IBMIR) warrant special awareness in clinical cell preparation for intravascular delivery to yield optimalperformance in vivo. Triggering of IBMIR leads to activation of the complement and coagulation systems leading to fibrin deposition,cell sequestration,embolization,anddamagein themicrovasculature. (D)Variousproduct innovationscanbeappliedto improvethesafety and efficacy of MSC products. Methods can either aim to suppress or abolish TF activity/presence. The first category entailsblocking substances (e.g., anti-TF mAb and iFVIIa), soluble anticoagulants (e.g., heparin), Ca2+ chelators (e.g., EDTA), and cellmembranemodification(e.g.,amacromolecularheparin-conjugatecoating) [44,50].Thesecondcategoryentailsminimalexpansioninculture(e.g.,BM-MSCs) [7],TF+/�cell/graftselection(e.g., throughmagnetic-orfluorescence-basedcellsorting) [22],orantisenseRNA(asRNA)againstTFtranscriptsandCRISPR/Cas9genomeediting,whichhavenotyetbeenexploredat large inexperimentalorclinical approaches. Abbreviations: CRISPR/Cas9, clustered regularly interspaced short palindromic repeats genome-editingsystem; EDTA, ethylenediaminetetraacetic acid; iFVIIa, inactivated activated coagulation factor VII; mAb, monoclonal antibody.

Clinician’s CornerMSC products have greatly diversifiedover recent years, and reports of safetyconcerns, severe adverse events, andeven fatal human cases caused bythrombotic complications with newintravascular MSC products highlightthe urgent clinical need for new stand-ards for hemocompatibility testing.

Although conventional BM-MSCs dis-play favorable hemocompatibility,novel MSC products derived fromother tissue sources (e.g., adiposeand perinatal tissue) express higherlevels of procoagulant TF/CD142 andinduce stronger triggering of IBMIR,which may compromise their safetyprofile upon intravascular delivery.

The ISCT minimal criteria for MSCscurrently do not assess or guaranteehemocompatibility because they wereinitially not designed for intravascularcell application. We propose herestandardization of hemocompatibilitytesting for MSC products intendedfor intravascular use.

To raise awareness within the clinical,scientific, regulatory, and patient com-munities, this review provides an over-view of this emerging safety issue, withexpert recommendations for stan-dardized hemocompatibility assess-ment, and the introduction ofinnovations in cell manufacturing andapplication procedures.

This perspective is in line with jointefforts by the FDA, Health Canada,ISCT, and other major representativesof academia, industry, bioethics, andpatient advocacy groups for a wideraction plan to better protect patients(https://www.celltherapysociety.org/news/news.asp?id=400602).


Outstanding QuestionsAre tissue-resident native MSCs com-patible with circulation in the blood,and for how long? How does thischange with ex vivo expansion? Forautologous and iPSC products, howdo patient attributes and comorbiditiescontribute to the hemocompatibility ofcell products?

What signaling pathways underlie theinduction of TF expression in differentcell types in vivo and in cell products exvivo, and can it be reduced or abol-ished by culture in advanced bioreac-tors, pharmacological treatment, orgenetic manipulation?

Which MSC products, dosing regi-mens, and delivery strategies offerthe best combination of safety andefficacy for specific target indications?What are the ‘minimal safety criteria’for MSC products intended for sys-temic intravascular delivery?

Should hemocompatibility assess-ment be added as release criteria forintravascular MSC therapeutics tocomplement the existing MSC minimalcriteria, and, if so, should it beassessed before or directly afterretrieval from cryopreservation?

Is the triggering of IBMIR by therapeu-tic cells necessary to realize MSC ther-apeutic benefit – and if MSCs andother cell products are modified toimprove hemocompatibility, how mightthis alter other aspects of cell function?

During pregnancy, several grams of fetal trophoblast-derived stromal material embolizes inthe lung microvasculature each day, locally reprogramming maternal lung macrophages byefferocytosis to promote systemic fetomaternal tolerance, thus making PT-derived MSCs anatural choice to recapitulate this phenomenon [1]. It is not clear if MSC TF expression andtriggering of IBMIR are a necessary or undesired component from a functional point of view,but it is clear from a safety perspective that optimal control of IBMIR is in the best interest ofthe patient.

Testing TF expression and in vitro hemocompatibility before infusion could help to address thisclinical need (Figure 5). Exclusion of TF-positive cells by cell-sorting strategies does notcompromise the immunomodulatory activity of the cells in vitro [23]. Methods to downregulateor eliminate TF expression by MSC products (e.g., by using CRISPR/Cas9 or antisense RNA),as well as sophisticated cell-surface coatings to promote anticoagulant and complement-inhibitory properties, are valuable alternatives to systemic anticoagulation [48]. Alternativemodes of delivery (e.g., intramuscular injection or extracorporeal) are also accessible[94,95]. Biomaterial-assisted delivery may overcome problems associated with inadequatehemocompatibility and provide an appropriate supportive environment to improve their trophicactivity in vivo [96].

Concluding Remarks and Future PerspectivesMSC products have greatly diversified during the past decade, complicating comparabilitybetween products, with potentially serious consequences. Clinical case reports of severeadverse events, such as infusion reactions and thromboembolism, after infusion of novelintravascular MSC products call for new safety standards. Although united under a singlename based on minimal criteria, there are significant functional differences between MSCproducts derived from different tissue sources. We have discussed here the molecular basis ofwhy distinct products show different propensities to trigger adverse innate immune responses,and we propose new methods to assess hemocompatibility before clinical use. Furthermore,we have discussed how optimized delivery or bioengineering of the cells themselves can beused to avoid or attenuate IBMIR. Hemocompatibility testing and optimal product delivery arecrucial for designing safer and more effective intravascular MSC therapies, paralleling otherexperimental intravascular therapies with cells that are not native to blood � such as islets,hepatocytes, and some products derived from induced pluripotent stem cells (iPSCs). Over thepast decades, MSC therapies have come a long way, and we must not lose sight of the fact thatMSC minimal criteria are indeed minimal, and the equivalence of new products cannot beassumed, and thus comprehensive safety evaluation is essential before use in humans. Recentclinical trial successes have demonstrated that by choosing a potent cellular therapeutic andoptimizing its delivery the clinical outcomes for patients can be greatly improved.

AcknowledgmentsG.M. was supported by funding from the German Research Foundation (DFG) and the German Federal Ministry of

Education and Research (BMBF) through the BSRT (GSC203) and BCRT. S.G. was supported by grants from the BMBF

(DIMEOS, 01EC1402B) and DFG (FOR2165, GE2512/2-2). This project has received funding from the EU Horizon 2020

Research and Innovation Programme under grant agreements 733006 (PACE) and 779293 (HIPGEN). J.A. was supported

by the Fraternal Order of Eagles Diabetes Research Center and the Straub Foundation. O.R. was supported by grants from

the Swedish Research Council (K2014-64X-05971-34-4), the Swedish Cancer Society (CAN2013/671), the Cancer

Society in Stockholm (111293), and the Karolinska Institutet.

Supplemental InformationSupplemental information associated with this article can be found online at https://doi.org/10.1016/j.molmed.2018.

12.006.


References
1. Galipeau, J. and Sensebe, L. (2018) Mesenchymal stromal cells:
clinical challenges and therapeutic opportunities. Cell Stem Cell22, 824–833

2. Mendicino, M. et al. (2014) MSC-based product characterizationfor clinical trials: an FDA perspective. Cell Stem Cell 14, 141–145

3. Ankrum, J. and Karp, J.M. (2010) Mesenchymal stem cell ther-apy: two steps forward, one step back. Trends Mol. Med. 16,203–209

4. Ankrum, J.A. et al. (2014) Mesenchymal stem cells: immuneevasive, not immune privileged. Nat. Biotechnol. 32, 252–260

5. Bianco, P. et al. (2013) The meaning, the sense and the signifi-cance: translating the science of mesenchymal stem cells intomedicine. Nat. Med. 19, 35–42

6. Griffin, M.D. et al. (2013) Concise review: adult mesenchymalstromal cell therapy for inflammatory diseases: how well are wejoining the dots? Stem Cells 31, 2033–2041

7. Moll, G. et al. (2012) Are therapeutic human mesenchymalstromal cells compatible with human blood? Stem Cells 30,1565–1574

8. Lalu, M.M. et al. (2012) Safety of cell therapy with mesenchymalstromal cells (SafeCell): a systematic review and meta-analysis ofclinical trials. PLoS One 7, e47559

9. Marks, P.W. et al. (2017) Clarifying stem-cell therapy’s benefitsand risks. N. Engl. J. Med. 376, 1007–1009

10. Cyranoski, D. (2010) Korean deaths spark inquiry. Nature 468,485

11. Jung, J.W. et al. (2013) Familial occurrence of pulmonary embo-lism after intravenous, adipose tissue-derived stem cell therapy.Yonsei Med. J. 54, 1293–1296

12. Wu, Z. et al. (2017) Thromboembolism induced by umbilical cordmesenchymal stem cell infusion: a report of two cases andliterature review. Transplant. Proc. 49, 1656–1658

13. George, M.J. et al. (2018) Clinical cellular therapeutics accelerateclot formation. Stem Cells Transl. Med. 7, 731–739

14. Vulliet, P.R. et al. (2004) Intra-coronary arterial injection of mes-enchymal stromal cells and microinfarction in dogs. Lancet 363,783–784

15. Furlani, D. et al. (2009) Is the intravascular administration ofmesenchymal stem cells safe? Mesenchymal stem cells andintravital microscopy. Microvasc. Res. 77, 370–376

16. Toma, C. et al. (2009) Fate of culture-expanded mesenchymalstem cells in the microvasculature: in vivo observations of cellkinetics. Circ. Res. 104, 398–402

17. Stephenne, X. et al. (2012) Bivalirudin in combination with heparinto control mesenchymal cell procoagulant activity. PLoS One 7,e42819

18. Tatsumi, K. et al. (2013) Tissue factor triggers procoagulation intransplanted mesenchymal stem cells leading to thromboembo-lism. Biochem. Biophys. Res. Commun. 431, 203–209

19. Gleeson, B.M. et al. (2015) Bone marrow-derived mesenchymalstem cells have innate procoagulant activity and cause microvas-cular obstruction following intracoronary delivery: amelioration byantithrombin therapy. Stem Cells 33, 2726–2737

20. Moll, G. et al. (2015) Different procoagulant activity of therapeuticmesenchymal stromal cells derived from bone marrow and pla-cental decidua. Stem Cells Dev. 24, 2269–2279

21. Shiratsuki, S. et al. (2015) Enhanced survival of mice infused withbone marrow-derived as compared with adipose-derived mes-enchymal stem cells. Hepatol. Res. 45, 1353–1359

22. Christy, B.A. et al. (2017) Procoagulant activity of human mesen-chymal stem cells. J. Trauma Acute Care Surg. 83, S164–S169

23. Oeller, M. et al. (2018) Selection of tissue factor-deficient celltransplants as a novel strategy for improving hemocompatibility ofhuman bone marrow stromal cells. Theranostics 8, 1421–1434

24. Liao, L. et al. (2017) Heparin improves BMSC cell therapy: anti-coagulant treatment by heparin improves the safety and thera-peutic effect of bone marrow-derived mesenchymal stem cellcytotherapy. Theranostics 7, 106–116

25. Perlee, D. et al. (2018) Intravenous infusion of human adiposemesenchymal stem cells modifies the host response to lipopoly-saccharide in humans: a randomized, single-blind, parallel group,placebo controlled trial. Stem Cells 36, 1778–1788

26. Dominici, M. et al. (2006) Minimal criteria for defining multipotentmesenchymal stromal cells. The International Society for CellularTherapy position statement. Cytotherapy 8, 315–317

27. Galipeau, J. et al. (2016) International Society for Cellular Therapyperspective on immune functional assays for mesenchymal stro-mal cells as potency release criterion for advanced phase clinicaltrials. Cytotherapy 18, 151–159

28. Friedenstein, A.J. et al. (1974) Stromal cells responsible for trans-ferring the microenvironment of the hemopoietic tissues. Cloningin vitro and retransplantation in vivo. Transplantation 17, 331–340

29. Pittenger, M.F. et al. (1999) Multilineage potential of adult humanmesenchymal stem cells. Science 284, 143–147

30. von Bahr, L. et al. (2012) Analysis of tissues following mesenchy-mal stromal cell therapy in humans indicates limited long-termengraftment and no ectopic tissue formation. Stem Cells 30,1575–1578

31. Horwitz, E.M. et al. (1999) Transplantability and therapeuticeffects of bone marrow-derived mesenchymal cells in childrenwith osteogenesis imperfecta. Nat. Med. 5, 309–313

32. Horwitz, E.M. et al. (2002) Isolated allogeneic bone marrow-derived mesenchymal cells engraft and stimulate growth in chil-dren with osteogenesis imperfecta: implications for cell therapy ofbone. Proc. Natl. Acad. Sci. U. S. A. 99, 8932–8937

33. Crisan, M. et al. (2008) A perivascular origin for mesenchymalstem cells in multiple human organs. Cell Stem Cell 3, 301–313

34. Bianco, P. et al. (2008) Mesenchymal stem cells: revisiting history,concepts, and assays. Cell Stem Cell 2, 313–319

35. Tsai, M.S. et al. (2007) Functional network analysis of the tran-scriptomes of mesenchymal stem cells derived from amnioticfluid, amniotic membrane, cord blood, and bone marrow. StemCells 25, 2511–2523

36. Parolini, O. et al. (2008) Concise review: isolation and characteri-zation of cells from human term placenta: outcome of the firstinternational Workshop on Placenta Derived Stem Cells. StemCells 26, 300–311

37. Kusuma, G.D. et al. (2015) Mesenchymal stem cells reside in avascular niche in the decidua basalis and are absent in remod-elled spiral arterioles. Placenta 36, 312–321

38. Vento-Tormo, R. et al. (2018) Single-cell reconstruction of theearly maternal-fetal interface in humans. Nature 563, 347–353

39. Ringden, O. et al. (2018) Placenta-derived decidua stromal cellsfor treatment of severe acute graft-versus-host disease. StemCells Transl. Med. 7, 325–331

40. Moll, G. et al. (2014) Do cryopreserved mesenchymal stromalcells display impaired immunomodulatory and therapeutic prop-erties? Stem Cells 32, 2430–2442

41. Witkowski, M. et al. (2016) Tissue factor as a link betweeninflammation and coagulation. Trends Cardiovasc. Med. 26,297–303

42. Drake, T.A. et al. (1989) Selective cellular expression of tissuefactor in human tissues. Implications for disorders of hemostasisand thrombosis. Am. J. Pathol. 134, 1087–1097

43. Faulk, W.P. et al. (1990) Tissue factor: identification andcharacterization of cell types in human placentae. Blood76, 86–96

44. Moberg, L. et al. (2002) Production of tissue factor by pancreaticislet cells as a trigger of detrimental thrombotic reactions in clinicalislet transplantation. Lancet 360, 2039–2045

45. Shapiro, A.M. et al. (2000) Islet transplantation in seven patientswith type 1 diabetes mellitus using a glucocorticoid-free immu-nosuppressive regimen. N. Engl. J. Med. 343, 230–238

46. Ryan, E.A. et al. (2001) Clinical outcomes and insulin secretionafter islet transplantation with the Edmonton protocol. Diabetes50, 710–719


47. Koh, A. et al. (2010) Insulin-heparin infusions peritransplant sub-stantially improve single-donor clinical islet transplant success.Transplantation 89, 465–471

48. Nilsson, B. et al. (2010) Can cells and biomaterials in therapeuticmedicine be shielded from innate immune recognition? TrendsImmunol. 31, 32–38

49. Karp, J.M. and Leng Teo, G.S. (2009) Mesenchymal stem cellhoming: the devil is in the details. Cell Stem Cell 4, 206–216

50. Kuznetsov, S.A. et al. (2007) Circulating connective tissue pre-cursors: extreme rarity in humans and chondrogenic potential inguinea pigs. Stem Cells 25, 1830–1839

51. He, Q. et al. (2007) Concise review: multipotent mesenchymalstromal cells in blood. Stem Cells 25, 69–77

52. Jones, E. and McGonagle, D. (2008) Human bone marrow mes-enchymal stem cells in vivo. Rheumatology (Oxford) 47, 126–131

53. Ekdahl, K.N. et al. (2015) Thromboinflammation in therapeuticmedicine. Adv. Exp. Med. Biol. 865, 3–17

54. Bennet, W. et al. (1999) Incompatibility between human bloodand isolated islets of Langerhans: a finding with implications forclinical intraportal islet transplantation? Diabetes 48, 1907–1914

55. Devine, S.M. et al. (2003) Mesenchymal stem cells distribute to awide range of tissues following systemic infusion into nonhumanprimates. Blood 101, 2999–3001

56. Fischer, U.M. et al. (2009) Pulmonary passage is a major obstaclefor intravenous stem cell delivery: the pulmonary first-pass effect.Stem Cells Dev. 18, 683–692

57. Asahara, T. et al. (1997) Isolation of putative progenitor endothe-lial cells for angiogenesis. Science 275, 964–967

58. Mendez-Ferrer, S. et al. (2010) Mesenchymal and haematopoieticstem cells form a unique bone marrow niche. Nature 466, 829–834

59. Pitchford, S.C. et al. (2009) Differential mobilization of subsets ofprogenitor cells from the bone marrow. Cell Stem Cell 4, 62–72

60. Marquez-Curtis, L.A. et al. (2011) The ins and outs of hemato-poietic stem cells: studies to improve transplantation outcomes.Stem Cell Rev. 7, 590–607

61. Schraufstatter, I.U. et al. (2009) C3a and C5a are chemotacticfactors for human mesenchymal stem cells, which cause pro-longed ERK1/2 phosphorylation. J. Immunol. 182, 3827–3836

62. Moll, G. et al. (2011) Mesenchymal stromal cells engage comple-ment and complement receptor bearing innate effector cells tomodulate immune responses. PLoS One 6, e21703

63. Mendez-Ferrer, S. et al. (2008) Haematopoietic stem cell releaseis regulated by circadian oscillations. Nature 452, 442–447

64. Thomas, H.E. et al. (2008) Circulating endothelial progenitorcells exhibit diurnal variation. Arterioscler. Thromb. Vasc. Biol.28, e21–e22

65. Hoogduijn, M.J. et al. (2014) No evidence for circulating mesen-chymal stem cells in patients with organ injury. Stem Cells Dev.23, 2328–2335

66. Alm, J.J. et al. (2010) Circulating plastic adherent mesenchymalstem cells in aged hip fracture patients. J. Orthop. Res. 28, 1634–1642

67. Teo, G.S. et al. (2012) Mesenchymal stem cells transmigratebetween and directly through tumor necrosis factor-alpha-acti-vated endothelial cells via both leukocyte-like and novel mecha-nisms. Stem Cells 30, 2472–2486

68. Wu, K.K. and Thiagarajan, P. (1996) Role of endothelium inthrombosis and hemostasis. Annu. Rev. Med. 47, 315–331

69. Nachman, R.L. and Jaffe, E.A. (2004) Endothelial cell culture:beginnings of modern vascular biology. J. Clin. Invest. 114, 1037–1040

70. Li, Y. and Lin, F. (2012) Mesenchymal stem cells are injured bycomplement after their contact with serum. Blood 120, 3436–3443

71. Steffel, J. et al. (2006) Tissue factor in cardiovascular diseases:molecular mechanisms and clinical implications. Circulation 113,722–731


72. Markiewski, M.M. et al. (2007) Complement and coagula-tion: strangers or partners in crime? Trends Immunol. 28,184–192

73. Ricklin, D. et al. (2010) Complement: a key system for immunesurveillance and homeostasis. Nat. Immunol. 11, 785–797

74. Ekdahl, K.N. et al. (2018) Interpretation of serological comple-ment biomarkers in disease. Front. Immunol. 9, 2237

75. Hashmi, S. et al. (2016) Survival after mesenchymal stromal celltherapy in steroid-refractory acute graft-versus-host disease:systematic review and meta-analysis. Lancet Haematol. 3,e45–52

76. Ringden, O. (2015) Mesenchymal stem (stromal) cells for treat-ment of acute respiratory distress syndrome. Lancet Respir. Med.3, e12

77. Lublin, F.D. et al. (2014) Human placenta-derived cells (PDA-001)for the treatment of adults with multiple sclerosis: a randomized,placebo-controlled, multiple-dose study. Multi Scler. Relat. Dis-ord. 3, 696–704

78. Melmed, G.Y. et al. (2015) Human placenta-derived cells (PDA-001) for the treatment of moderate-to-severe Crohn’s disease: aPhase 1b/2a study. Inflamm. Bowel Dis. 21, 1809–1816

79. Baygan, A. et al. (2017) Safety and side effects of using placenta-derived decidual stromal cells for graft-versus-host disease andhemorrhagic cystitis. Front. Immunol. 8, 795

80. Liang, J. et al. (2018) Safety analysis in patients with autoimmunedisease receiving allogeneic mesenchymal stem cells infusion: along-term retrospective study. Stem Cell Res. Ther. 9, 312

81. Hoogduijn, M.J. et al. (2016) Effects of freeze-thawing and intra-venous infusion on mesenchymal stromal cell gene expression.Stem Cells Dev. 25, 586–597

82. Acosta, L. et al. (2013) Adipose mesenchymal stromal cellsisolated from type 2 diabetic patients display reduced fibrinolyticactivity. Diabetes 62, 4266–4269

83. Toyserkani, N.M. et al. (2017) Concise review: a safety assess-ment of adipose-derived cell therapy in clinical trials: a systematicreview of reported adverse events. Stem Cells Transl. Med. 6,1786–1794

84. Capilla-Gonzalez, V. et al. (2018) PDGF restores the defectivephenotype of adipose-derived mesenchymal stromal cells fromdiabetic patients. Mol. Ther. 26, 2696–2709

85. Davies, L.C. et al. (2016) Type 1 diabetes mellitus donor mesen-chymal stromal cells exhibit comparable potency to healthy con-trols in vitro. Stem Cells Transl. Med. 5, 1485–1495

86. Moll, G. et al. (2014) Do ABO blood group antigens hamper thetherapeutic efficacy of mesenchymal stromal cells? PLoS One 9,e85040

87. Deak, E. et al. (2010) Suspension medium influences interac-tion of mesenchymal stromal cells with endothelium and pul-monary toxicity after transplantation in mice. Cytotherapy 12,260–264

88. Quimby, J.M. et al. (2013) Safety and efficacy of intravenousinfusion of allogeneic cryopreserved mesenchymal stem cellsfor treatment of chronic kidney disease in cats: results of threesequential pilot studies. Stem Cell Res. Ther. 4, 48

89. Gustafson, E. et al. (2017) Control of IBMIR induced by fresh andcryopreserved hepatocytes by low molecular weight dextransulfate versus heparin. Cell Transplant. 26, 71–81

90. Netsch, P. et al. (2018) Human mesenchymal stromal cells inhibitplatelet activation and aggregation involving CD73-convertedadenosine. Stem Cell Res. Ther. 9, 184

91. Moll, G. and Le Blanc, K. (2015) Engineering more efficient multi-potent mesenchymal stromal (stem) cells for systemic delivery ascellular therapy. ISBT Sci. Ser. 10, 357–365

92. Moll, G. et al. (2016) Cryopreserved or fresh mesenchymal stro-mal cells: only a matter of taste or key to unleash the full clinicalpotential of MSC therapy? Adv. Exp. Med. Biol. 951, 77–98

93. Aronsson-Kurttila, W. et al. (2018) Placenta-derived deciduastromal cells for hemorrhagic cystitis after stem cell transplanta-tion. Acta Haematol. 139, 106–114

94. Struecker, B. et al. (2014) Liver support strategies: cutting-edge technologies. Nat. Rev. Gastroenterol. Hepatol. 11,166–176

95. Miller, B.L.K. et al. (2018) Extracorporeal stromal cell therapy forsubjects with dialysis-dependent acute kidney injury. Kidney Int.Rep. 3, 1119–1127

96. Qazi, T.H. et al. (2015) Biomaterials based strategies for skeletalmuscle tissue engineering: existing technologies and futuretrends. Biomaterials 53, 502–521

97. Soland, M.A. et al. (2013) Mesenchymal stem cells engi-neered to inhibit complement-mediated damage. PLoS One8, e60461

98. Li, Y. et al. (2016) Painting factor H onto mesenchymal stem cellsprotects the cells from complement- and neutrophil-mediateddamage. Biomaterials 102, 209–219

99. Li, Y. et al. (2016) Local inhibition of complement improvesmesenchymal stem cell viability and function after administration.Mol. Ther. 24, 1665–1674


Review Intravascular Mesenchymal Stromal/Stem Cell Therapy ... · (BIH), Berlin, Germany...

Documents

Transcript of Review Intravascular Mesenchymal Stromal/Stem Cell Therapy ... · (BIH), Berlin, Germany...