Cardiac repair with allogeneic mesenchymal stem cells after myocardial infarction

Over the past decade, use of autologous bone marrow-derived mononuclear cells (BMCs) has proven to be safe in phase-I/II studies in patients with myocardial infarction (MI). Taken as a whole, results support a modest yet significant improvement in cardiac function in cell-treated patients. Skeletal myoblasts, adipose-derived stem cells, and bone marrow-derived mesenchymal stem cells (MSCs) have also been tested in clinical studies. MSCs expand rapidly in vitro and have a potential for multilineage differentiation. However, their regenerative capacity decreases with aging, limiting efficacy in old patients. Allogeneic MSCs offer several advantages over autologous BMCs; however, immune rejection of allogeneic cells remains a key issue. As human MSCs do not express the human leukocyte antigen (HLA) class II under normal conditions, and because they modulate T-cell-mediated responses, it has been proposed that allogeneic MSCs may escape immunosurveillance. However, recent data suggest that allogeneic MSCs may switch immune states in vivo to express HLA class II, present alloantigen and induce immune rejection. Allogeneic MSCs, unlike syngeneic ones, were eliminated from rat hearts by 5 weeks, with a loss of functional benefit. Allogeneic MSCs have also been tested in initial clinical studies in cardiology patients. Intravenous allogeneic MSC infusion has proven to be safe in a phase-I trial in patients with acute MI. Endoventricular allogeneic MSC injection has been associated with reduced adverse cardiac events in a phase-II trial in patients with chronic heart failure. The long-term safety and efficacy of allogeneic MSCs for cardiac repair remain to be established. Ongoing phase-II trials are addressing these issues.


Introduction
Growing evidence suggests that the adult mammalian heart harbours stem and progenitor cells that possess a regenerative capacity [1].Moreover, circulating bone marrow-derived progenitor cells can home to injured myocardium to participate in cardiac repair [2].Nevertheless, myocardial infarction (MI) results in a massive loss of cardiomyocytes and scar formation.Hence, endogenous regenerative mechanisms appear to be insufficient to replace the huge number of cardiomyocytes that are lost after MI.Chronic heart failure (CHF) eventually ensues in many patients.Recent advances in pharmacological agents, coronary revascularisation and implantable devices have succeeded in alleviating clinical symptoms, prolonging survival and slowing down the progression of CHF; however, they have failed to fully prevent CHF.As a loss of cardiomyocytes is primarily responsible for the development of CHF, the generation of new functional cardiomyocytes remains the ultimate goal for preserving cardiac function.Ten years ago, a seminal report claimed that adult mouse bone marrow-derived stem cells trans-differentiated into cardiomyocytes to regenerate infarcted myocardium in vivo [3].This report sparked overwhelming enthusiasm in the scientific community, leading to the extremely rapid initiation of clinical studies of bone marrow cell (BMC)based therapies for cardiac repair [4,5].Meanwhile, cardiomyogenic trans-differentiation of mouse BMCs was not confirmed by subsequent studies from other laboratories [6, 7], questioning the very rationale for the initiation of these clinical studies.The first phase-I study of BMCbased therapy for ischemic heart disease [5] was published only 2 years after the initial experimental report on cardiomyogenic trans-differentiation of BMCs [3].Several other phase-I/II clinical studies were published in the next few years.Most of these studies adopted a "blanket" strategy using autologous unfractionated mononuclear BMCs, rather than selected cell components within the heterogeneous BMC population.A few days after acute MI and percutaneous recanalisation of the infarct-related coronary artery, bone marrow was taken from the patient, and mononuclear BMCs were prepared in the laboratory and immediately infused into the infarct-related vessel using a percutaneous over-the-wire delivery-catheter.Autologous BMCbased cell therapies for cardiac repair have proven to be safe in randomised controlled trials [8][9][10][11][12].In agreement with these studies, our own data from the phase-I Stem cell Transplantation in Ischemic Myocardium (STIM) trial (n = 23 patients) and preliminary data from the subsequent phase-II SWiss multicentre Intracoronary Stem cells Study in Acute Myocardial Infarction (SWISS-AMI) trial [13] (n = 166 patients enrolled as of March 31, 2011) indicate that intracoronary autologous BMC infusion in patients with acute MI is safe (Corti R, Sürder D, Moccetti T et al.; unpublished data).Efficacy data from randomised controlled trials have been mixed [8][9][10][11][12].Taken as a whole, these data support a small yet significant increase in left ventricular ejection fraction (EF) in cell-treated patients [14].Functional improvement seemed to be higher in patients with moderate-severe ventricular dysfunction than in those with mild dysfunction [10,12].However, in many instances, functional improvement was transient [15].Autologous mononuclear BMC-based therapies have several limitations.The cell component within the heterogeneous BMC population that accounts for functional benefit has not been precisely defined.Bone marrow stem cells represent only 1 in 10,000 BMCs [16], and therefore can hardly explain the beneficial effect of BMC transplantation.Also, the proliferative and regenerative capacity of progenitor cells decreases with aging [17,18], consistent with a reduced efficacy of progenitor cells from old patients.Another practical limitation is that the collection of bone marrow from the patient requires an invasive procedure.Skeletal myoblasts, adipose-derived stem cells and mesenchymal stem cells (MSCs) have also been tested in a limited number of clinical studies in patients with MI.Skeletal myoblasts form contractile striated myocytes that do not couple electromechanically with cardiomyocytes.Percutaneous intramyocardial injection of autologous skeletal myoblasts using a cell delivery-catheter has proven safe but ineffective in the recent SEISMIC trial in patients with CHF [19].Intracoronary infusion of autologous adiposederived stem cells was proven to be safe in the recent double-blind, placebo-controlled APOLLO trial in patients with acute MI [20].The cell-treated group showed a 5.7% increase in ejection fraction (EF) and improved myocardial perfusion, although these effects did not reach statistical significance in this early trial.The efficacy of adiposederived stem cells will be evaluated in the phase-II/III ADVANCE trial.The present short review aims to discuss MSC-based therapies after MI, with a focus on allogeneic MSC products.

Experimental studies of autologous MSCs for cardiac repair
Bone marrow MSCs are multipotent stromal (non haematopoietic) cells capable of differentiating into multiple lineages of the mesenchyme to become fat, bone, tendon and muscle.Cultured MSCs grow rapidly in vitro.Systemically delivered MSCs can home to sites of ischemia and tissue injury, including the ischemic heart.This remarkable property allows the non-invasive administration of MSCs intravenously, thus avoiding the need for intracoronary or intramyocardial cell delivery.MSCs treated with the demethylating agent 5-azacytidine or cardiomyogenic growth factors can adopt cardiac phenotypes in vitro [21][22][23][24][25].In many instances, however, cardiac differentiation has been limited to the expression of cardiac-specific genes, without generation of functional cardiomyocytes [26].Immunohis-tochemical evidence of cardiomyogenic differentiation of human MSCs implanted into infarcted mouse hearts has been reported [27].Several experimental studies have shown improvements in cardiac function after MI in animals receiving MSCs [28][29][30][31][32][33] despite infrequent differentiation [29,30].In contrast, no functional benefit was observed in a minority of studies [34].Over-expression of pro-survival molecules (e.g.Akt) by autologous MSCs significantly enhanced functional recovery after MI [30,35].These observations support central roles of paracrine mechanisms that stimulate angiogenesis and endogenous cardiac stem cells [36], rather than cell differentiation, in the therapeutic effects of MSCs.

Clinical studies of autologous MSCs for cardiac repair
Intracoronary infusion of autologous bone marrow-derived MSCs significantly increased EF at 3 months in a randomised study in patients with acute MI [37].In addition, autologous MSC transplantation has proven safe in several nonrandomised studies in patients with acute MI or CHF [38,39] As MSCs are rare among BMCs, they need to be expanded ex vivo for autologous applications.This procedure may take between 10 days [37] to 4-5 weeks [40] or longer depending on the cell number needed, and the age and comorbidities of the donor, as progenitor cells from old patients have reduced proliferation potential [17,18,41].Cell expansion requires a Good Manufacturing Practice (GMP) cell therapy unit (Figure 1).Allogeneic MSCs from young donors have several advantages over autologous MSCs, in- Good Manufacturing Practice (GMP) cell therapy unit at the Cardiocentro Ticino, Lugano.GMP facilities are required for ex vivo expansion of cells used in clinical studies.

Review article: Current opinion
Swiss Med Wkly.2011;141:w13209 cluding a high regenerative potential and availability for clinical use at any time point.In addition, no bone marrow collection from the patient is required.Moreover, MSCs from a single donor can be used for many patients, thus tremendously reducing the manufacturing costs of the cell product.On the other hand, the potential immunogenicity of allogeneic MSCs remains an unresolved issue that may affect clinical efficacy.

MSCs can switch immune states
Bone marrow-derived MSCs secrete immunoregulatory cytokines (e.g., IL10) and inhibit lymphocyte proliferation, dendritic cell maturation and alloimmune rejection [42][43][44][45][46][47][48].Due to their immunomodulatory properties, MSCs have been tested in initial clinical studies of graftversus-host diseases [49], osteogenesis imperfecta [50], glycogen storage diseases [51], Crohn's disease [52] and organ transplantation [53].MSCs have been shown to express major histocompatibility complex (MHC) molecules, including MHC class II (MHC-II).Expression of MHC molecules is regulated by pro-inflammatory cytokines.Adult human MSCs express intermediate levels of MHC-I but do not express human leukocyte antigen (HLA) class II or co-stimulatory molecules on the cell surface under normal in vitro conditions.However, HLA-II has been detected in lysates of unstimulated MSCs, presumably reflecting intracellular deposits of the antigen.Treatment with interferon-γ was found to induce expression of HLA class II on the cell surface [43].Another study showed that MSCs express MHC-II as antigen presenting cells in the presence of low levels of interferon-γ, whereas MHC-II expression was decreased by high levels of interferon-γ [54].Inducing MSCs to acquire cardiac phenotypes by 5-azacytidine or cytokine treatment increased expression of immunostimulatory MHC-Ia and MHC-II, while decreasing that of immunomodulatory MHC-IIb, resulting in enhanced cytotoxicity of MSCs in co-culture with allogeneic leukocytes [55].These observations suggest that MSCs can switch immune states from immunomodulatory to immunostimulatory (figure 2).Several studies have shown delayed host responses to al- logeneic MSCs in vivo around 2 weeks after transplantation, suggesting that tolerance of MSCs across the allogeneic barrier may not be absolute [55][56][57][58].

Experimental studies of allogeneic MSCs for cardiac repair
Intravenous [59] or intracoronary infusion, as well as transendocardial [60] or intramyocardial injection [28,61,62] of allogeneic MSCs has resulted in cell engraftment, reduced scar formation and improved cardiac function in various animal models of MI.Male MSCs injected into infarct scars and border zones in female allogeneic recipients differentiated towards multiple cell types including cardiomyocytes, vascular muscle cells and endothelial cells, as evidenced by co-localisation of Y-chromosome-positive cells and tissue-specific markers [62].MSC-derived cardiomyocytes integrated into host myocardium, forming gap junctions.In sheep, intracoronary infusion of allogeneic mesenchymal progenitor cells (MPCs) improved EF and blood vessel formation 2 months after MI (Henry T et al.; www.mesoblast.com).In a rat model, both allogeneic and syngeneic MSCs improved EF 3 months after MI; however, allogeneic cells were eliminated from the heart by 5 weeks and functional benefit was lost within 5 months [55].Circulating antibodies against differentiated allogeneic MSCs, but not against undifferentiated MSCs, were detected in the recipient by 5 weeks.In contrast, syngeneic MSCs were not eliminated and functional benefit lasted for more than 5 months.In a pig model, allogeneic MSCs were not immunogenic in vitro but elicited immune responses after intracardiac injection in vivo [58].These experimental findings indicate that bone marrow-derived MSCs can induce immune responses in allogeneic hosts.

Clinical studies of allogeneic MSCs for cardiac repair
The results of the first phase-I, randomised, double-blind, placebo-controlled, dose-escalation study of intravenous allogeneic adult MSCs in patients with acute MI were published in December 2009 [63].This study tested an "offthe-shelf" cell product (Prochymal TM , Osiris Tx.) made from MSCs of a single healthy donor.A total of 53 patients who had suffered a first MI in the previous 10 days were randomised to receive one of three intravenous doses of Prochymal TM or placebo, without immunosuppression.The study's primary end point, adverse event rates at 6 months, was similar in the cell-treated group and the placebo group, indicating that the procedure was safe.No procedure-related complications were observed.Renal, hepatic and hematologic laboratory indexes were similar in the two groups.No patient developed tumours.Cell-treated patients had less episodes of ventricular tachycardia and a better lung function compared with controls.Although the study did not primarily aim to assess clinical efficacy, there was a trend towards improved clinical symptoms and increased EF at 3 months, but not at 6 months, in the celltreated group.A phase-II multicentre trial of Prochymal TM has been started.Endoventricular injection of an allogeneic MPC product (Revascor TM , Mesoblast Ltd.) along the infarct border zone in patients with moderate-severe CHF has been evaluated in a phase-II, randomised, single-blind, placebo-controlled, dose-escalation, multicentre study.There were no procedure-related complications.The first cohort in the study (n=20 patients), receiving the low dose of Revascor TM in the trial, showed a 33% increase in EF at 6 months (p=0.029),whereas EF decreased 6% in the control group [64].Interim data analysis after all patients had reached 6 months follow-up (www.mesoblast.com/newsroom/asx-announcements/archives)showed that the number of patients who developed any severe adverse cardiac events over the follow-up period was 44.4% in the celltreated group (n=45) versus 93.3% in the control group (n = 15; p = 0.001).The number of patients who developed any major adverse cardiac events, defined as the composite of cardiac death, heart attack or coronary revascularisation procedures, was 6.7% in the cell-treated group versus 40% in the control group (p = 0.005).

Conclusions
"Off-the-shelf" allogeneic MSC products have proven safe in phase-I/II trials in patients with acute MI or CHF [63,64].Preliminary efficacy data are encouraging.Several issues remain unresolved, however.Little is known regarding the in vivo survival of allogeneic MSCs.Experimental evidence suggests allogeneic MSCs may switch immune states in vivo, inducing alloimmune rejection [55].The long-term safety and efficacy of allogeneic MSC-based therapies for cardiac repair are currently being tested in phase-II clinical trials.If immunogenicity of human allogeneic MSCs turned out to limit the clinical efficacy, pharmacological immunosuppression for limited periods of time after cell transplantation might be worth considering as a possible option, although the effectiveness of this approach remains to be demonstrated.Moreover, minimizing donor-recipient HLA-II mismatch might attenuate alloimmune rejection of transplanted MSCs.On the other hand, paracrine mechanisms of action [35] suggest that long-term survival of transplanted MSCs might not even be required to achieve sustained clinical effects.

Figure 2
Figure 2Cartoon depicting MSC immune switch.Under normal in vitro conditions (left), human MSCs express low MHC-I and no HLA class II, while secreting IL10.They mediate immunomodulatory effects on T cells.In the presence of pro-inflammatory cytokines (right), HLA class II is expressed on cell surfaces.MSCs acquire antigen presenting cell properties and stimulate T cells.