Prospects for stem cell transplantation in the treatment of hepatic disease


  • Benjamin M. Stutchfield,

    Corresponding author
    1. MRC Centre for Inflammation Research and MRC Centre for Regenerative Medicine, University of Edinburgh, Edinburgh, United Kingdom
    • c/o Stephen J. Wigmore, MD, FRCS, Medical Research Council Centre for Inflammation Research and Medical Research Council Centre for Regenerative Medicine, Queen's Medical Research Institute, University of Edinburgh, 47 Little France Crescent, Edinburgh, EH16 4TJ, United Kingdom
    Search for more papers by this author
    • Telephone: +44 131 242 1000

  • Stuart J. Forbes,

    1. MRC Centre for Inflammation Research and MRC Centre for Regenerative Medicine, University of Edinburgh, Edinburgh, United Kingdom
    Search for more papers by this author
  • Stephen J. Wigmore

    1. MRC Centre for Inflammation Research and MRC Centre for Regenerative Medicine, University of Edinburgh, Edinburgh, United Kingdom
    Search for more papers by this author


Stem cell therapy has the potential to provide a valuable adjunct to the management of hepatic disease. Preclinical studies have demonstrated a range of endogenous repair processes that can be exploited through stem cell therapy. Initial translational studies have been encouraging and have suggested improved liver function in advanced chronic liver disease and enhanced liver regeneration after portal vein embolization. This article reviews the potential for stem cell therapies to enhance hepatic regeneration in acute and chronic hepatic disease and is based on a MEDLINE and PubMed search for English language articles investigating mechanisms of hepatic regeneration and delivery of cell therapies. Two main mechanisms of potential stem cell therapy delivery have emerged: (1) a direct contribution to the functional hepatocyte population with embryonic, induced pluripotent, or adult stem cells and (2) the promotion of endogenous regenerative processes with bone marrow–derived stem cells. Bioartificial hepatic support systems may be proven to be an effective method of using ex vivo differentiated hepatocytes and be indicated as a bridging therapy to definitive surgery in acute liver failure. The administration of bone marrow–derived stem cells may enhance liver regeneration in chronic liver disease after portal vein embolization and could facilitate regeneration after partial hepatic resection. Ultimately, the most appropriate hepatic disease targets for stem cell therapies will become apparent as mechanisms of stem involvement in hepatic regeneration are further elucidated. Liver Transpl 16:827–836, 2010. © 2010 AASLD.

Stem cell therapies offer the opportunity to transform the approach to hepatic disease. Although current pharmacological therapy may target specific pathways or receptors, stem cell therapies can provide a living agent able to influence a range of biological processes. The potential to enhance hepatic regeneration after partial hepatectomy in advanced cirrhosis or provide a bridge to orthotopic liver transplantation has focused research.1-3 Hepatocyte-like cultures have been generated in vitro from both embryonic stem cells (ESCs) and human peripheral blood monocytes,4, 5 and the direct administration of cell therapies in rodent models has been shown to support hepatic function.5, 6 Initial translational pilot studies testing the direct hepatic administration of bone marrow–derived stem cells have been encouraging and have suggested enhanced liver regeneration prior to partial hepatectomy and improved liver function in advanced chronic liver disease.7-11

With rapidly mounting preclinical and clinical evidence, the delivery of stem cell therapies to support hepatic function may now be realistically contemplated. The development of potential stem cell therapies as an adjunct to the management of hepatic disease requires a careful and rational evaluation of the preclinical studies and a comprehensive review of the initial clinical investigations. The purpose of this review is to address the potential for cell therapies to enhance hepatic regeneration and the practical barriers currently restricting their development.


BALS, bioartificial liver support; BMSC, bone marrow stem cell; ESC, embryonic stem cell; G-CSF, granulocyte colony stimulating factor; HSC, hematopoietic stem cell; INR, international normalized ratio; IPSC, induced pluripotent stem cell; MELD, Model for End-Stage Liver Disease; MSC, mesenchymal stem cell; N/A, not applicable.


Hepatic regeneration involves a complex interaction of cytokines, growth factors, resident hepatic cells, and bone marrow–derived cells.12 In response to a regenerative stimulus such as partial hepatectomy, the majority of normally quiescent hepatocytes enter the cell cycle.12 In severe acute or chronic hepatic injury, the regenerative capacity of resident hepatocytes may be overwhelmed, and a flexible, multitiered system of progenitor or stem cells is invoked to support hepatic regeneration.13

Adult tissue–specific stem cells exist within distinct regions, stem cell niches, which provide a specialized microenvironment for the regulation and maintenance of stem cells. The exact location of the liver stem cell niche in both rodents and humans is the source of much debate. In rodents, these stem cells are termed oval cells and are capable of differentiation into both hepatocytes and biliary epithelia. Oval cells have long been thought to exist in the terminal branches of the biliary tree (the canals of Hering), but other stem cell niches may exist (intralobular bile ducts, periductal mononuclear cells, and peribiliary hepatocytes).13, 14 These findings likely reflect an intricate system of liver stem cells rather than a single hepatic stem cell location.

In the injured human, the small, heterogeneous nature of the progenitor cell fraction has, until recently, made these cells difficult to identify. Using immunohistochemical techniques, Fellous et al.15 were able to identify putative clonal proliferative units in the adult human liver often adjacent to the portal area. It has been proposed that these cells originate in the portal area and then migrate toward the hepatic veins. This observation has previously been described as the streaming liver hypothesis: cells generated at the periportal area represent one end of the flux and migrate down a path leading to the hepatic vein.16 This hypothesis is based on the finding that labeled hepatocytes in adult rats migrate from the portal rim to the central vein.16, 17 This is controversial, however, with others failing to demonstrate such a regenerative relationship with vascular structures.18-20 The examination of human fetal livers has identified a further population of stem cells termed side population cells, which are able to contribute to both hematopoietic and epithelial cell lineages and potentially contribute to hepatocyte generation.21 The presence of these cells in the adult liver and their ability to contribute to hepatic regeneration are uncertain.

After a hepatic injury such as partial hepatectomy, inflammatory liver disease, or the ischemic insult of orthotopic liver transplantation, bone marrow–derived hematopoietic stem cells (HSCs) are mobilized.22-24 Liver injury up-regulates the hepatic production of the potent HSC chemoattractant stromal cell–derived factor 1, and this provides a homing mechanism focused on the injured liver.24, 25 A range of growth factors (interleukin 8, matrix metalloproteinase 9, hepatocyte growth factor, and stem cell factor) also show increased expression after hepatic injury and contribute to HSC homing and engraftment.26 It has previously been suggested that bone marrow–derived stem cells are able to transdifferentiate into hepatocytes, and initial rodent studies demonstrating functional improvement appeared to support this hypothesis.27-29 However, these early studies have not been repeatable. It appears that either bone marrow–derived hepatocytes were erroneously identified or animal models with unique selection pressures were used that are not directly relevant to clinical practice.30 Bone marrow–derived stem cells are now not thought to be able to transdifferentiate into a clinically relevant quantity of hepatocytes but rather facilitate hepatic regeneration by supporting resident hepatocyte functions that promote vascular remodeling, macrophage-led matrix remodeling, and immune modulation. Two main cellular regenerative processes available to the injured liver have therefore emerged: (1) the proliferation of resident hepatocytes and hepatocyte progenitor cells and (2) the supportive function of bone marrow–derived stem cells. Stem cell therapies may exploit these 2 mechanisms by focusing either on (1) directly contributing to the hepatocyte population or (2) using bone marrow–derived stem cells to promote endogenous processes (Table 1).

Table 1. Autologous Stem Cell Therapies and Clinical Studies in Liver Disease
 Cell SourceIntended MechanismEvidenceComments
Contributing to the hepatocyte population
 Direct cell administration
  Undifferentiated   cellsESCsHepatic engraftment and hepatocyte differentiation and functionEngraftment and differentiation demonstrated but limited hepatocyte function6, 33, 34, 36Risk of malignancy with ESCs
Adult stem/progenitor cellsClinical trials unrealistic currently
Concerns regarding phenotypic stability
  Ex vivo   differentiated   hepatocytesESCsHepatic engraftment and hepatocyte functionLow level of engraftment with limited cell function35, 36, 54Risk of malignancy with ESCs
Adult stem/progenitor cellsLimited function of ex vivo differentiated hepatocytes at present
IPSCsClinical trials unrealistic currently
 BALS systemsEx vivo differentiated hepatocytesExtracorporeal hepatocytes contact-infuse blood via a semipermeable membrane.Improved survival in fulminant/ subfulminant hepatic failure with porcine-derived hepatocytes64Further advances required in efficient hepatocyte generation from human cells
Most potential in this field
Promoting endogenous processes
 Bone marrow–  derived stem  cellsHSCsMatrix remodeling, vascular remodeling, immunomodulation, and facilitation of resident hepatocyte differentiationImproved physiological parameters in phase 1 clinical trials8, 10, 11, 89, 95-98Therapeutic potential demonstrated
Further cautious clinical investigation required
Bone marrow MSCsMatrix remodeling, immunomodulation, and facilitation of resident hepatocyte differentiationImproved physiological parameters in a phase 1 clinical trial99Limited clinical evidence
MSC isolation problematic
Potential for increased hepatic fibrosis


Ideally, allogenic hepatocytes, ex vivo derived hepatocytes, or cells capable of hepatocyte differentiation could be administered directly and repopulate the failing liver. Allogenic hepatocyte transplantation has been explored as an alternative to orthotopic liver transplantation in acute liver failure and metabolic liver disease. However, difficulties in harvesting and storing sufficient quantities of hepatocytes and significant cell loss following transplantation have so far limited the potential of this therapy.31 A plentiful and reliable cell source is required.

Given the right environment and stimuli, stem cells and certain progenitor cells can differentiate into hepatocytes. Stem cells are undifferentiated cells capable of proliferation, self-maintenance, and differentiation into functional progeny with flexibility or plasticity in these options.32 ESCs have demonstrated pluripotency and have shown unlimited capacity for self-renewal. In contrast, adult stem cells tend to demonstrate a limited differentiation capability with single or defined target cells. Because of this, adult stem cells may be more correctly called progenitor cells. Despite the apparently limited differentiation capability of progenitor cells, given appropriate stimuli, progenitor cells can transdifferentiate into other cell lines.32 The ideal cell source to support hepatic regeneration must be reliably identifiable, be able to generate hepatocytes efficiently, evade the immune defenses, and behave predictably with a high safety profile.


ESCs, isolated from the inner cell mass of blastocyst-stage embryos, possess the most potent differentiation potential, with their capacity for self-renewal theoretically providing an unlimited supply of hepatocytes to support regeneration of the injured liver. In vitro differentiation of hepatocytes from ESCs is well documented, with protocols paralleling sequential hepatic development efficiently generating functional but immature hepatocytes.4, 33 When they are transplanted into rodent models of toxin-induced hepatic injury and partial hepatectomy, there is evidence of engraftment and differentiation into hepatocyte-like cells with some contribution to regeneration, but generally at low levels with minimal hepatocyte function.4, 6, 34, 35 As protocols improve, reports of efficient liver repopulation by fetal liver stem/progenitor cells are emerging.36 In comparison with the transplantation of adult hepatocytes, however, fetal liver progenitors and ESC-derived hepatic precursors currently appear less efficient at generating liver tissue in vivo.37 There are continuing efforts to improve the ability to generate, purify, and amplify ESC-derived hepatocyte-like cells with certain signaling proteins (activin A and wingless-type MMTV integration site family member 3A) that have been shown to improve the efficiency of differentiation.38

Exploiting the therapeutic potential of ESCs in a clinical setting presents a number of challenges. The ethical dilemmas in the use of human ESCs may continue to prohibit research in some societies, but there has been a gradual relaxation around the world as the therapeutic possibilities are realized. The potential for teratoma formation is likely to remain a concern until long-term trials can provide evidence of phenotypic stability and safety.39, 40 However, the major focus in contemplating clinical applications of ESCs is overcoming the immune barrier.

The immune response to transplanted organs is relatively well understood, but because ESC transplants lack donor-type vasculature, endothelial cells, and the antigen-presenting cells of transplanted organs, the course and intensity of the immune response are less clear. ESCs do display a certain degree of immune privilege because of the relatively low expression of major histocompatibility complex class I and II molecules, but they remain susceptible to rejection.41 Although thoughts of a human ESC bank may seem premature at present, it has been estimated that the number of ESC cell lines required to make ESC therapy accessible to a significant proportion of the population is approximately 150.42 This estimate is based only on selected major histocompatibility complex loci in combination with the judicious use of immunosuppression. However, expression of even minor histocompatibility antigens has been shown to be sufficient to induce acute rejection in tissues differentiated from ESCs.43 The processes and artificial manipulations associated with in vitro directed differentiation may themselves present novel targets for immune attack.44 No barrier is necessarily insurmountable, however, with the development of ESC culture techniques minimizing or eliminating the use of potentially immunogenic agents and ESC tolerance demonstrated in rodent models with minimal host conditioning.43, 45

The recently discovered ability to dedifferentiate adult somatic cells to a pluripotent state raises the possibility that stem cell therapies could be provided in a patient-specific manner potentially avoiding the immunological pitfalls. Such cells are termed induced pluripotent stem cells (IPSCs) and may be proven to be a promising tool for the delivery of regenerative stem cell therapies.


Somatic cells can be reprogrammed by the forced expression of transcriptional factors to a pluripotent state, and this creates IPSCs. This pluripotent state resembles ESCs in many aspects, including the expression of stem cell markers, the potential for teratoma formation, and the differentiation capacity.46 As with ESCs, hepatocyte-like cells have been efficiently generated in vivo from IPSCs derived from human fibroblasts.47, 48 Successful application in rodent models of disease is awaited. Although IPSCs generated from autologous sources have the potential to avoid rejection, the feeding layers involved in their culture may still contain potentially immunogenic material. Steps to produce clinical-grade IPSCs are well underway.49

Adult Stem/Progenitor Cells

Hepatocyte-like cells can also be generated directly from adult stem/progenitor cells, and this provides a potential advantage over both ESCs and IPSCs. Adult stem/progenitor cells avoid many of the ESC ethical issues and, in comparison with both ESCs and IPSCs, show no risk of teratoma formation. The potential to deliver autologous cells may again negate the requirement for immunosuppression. With resident human hepatic progenitor cells shown to possess mesenchymal stem cell (MSC) markers,50 multiple in vitro studies have demonstrated transdifferentiation of MSCs harvested from diverse sources into hepatocyte-like cells.51-53 Initial animal studies focused on the potential of MSCs to generate hepatocytes in vivo.54, 55 Although the generation of functional hepatocyte-like cells after direct hepatic administration was observed, the propensity of MSCs to form myofibroblast-like cells in areas of hepatic injury is a concern.56

Delivering Hepatocyte Stem Cell Therapies

Given the range of hepatocyte generation methods available, attention has been focused on finding the most appropriate niche in which ex vivo differentiated hepatocytes can function effectively. Certainly, direct hepatocyte administration has shown promise in treating metabolic disease, in which a low percentage of cell engraftment can still result in clinical improvement.57 In acute liver failure, the presence of strong regenerative demand may provide favorable conditions under which liver repopulation by exogenous cell lines can be facilitated.58, 59 However, in the presence of cirrhosis, the severely disrupted liver architecture may present a hostile target for effective cell engraftment, differentiation, and function. Using alternative sites to deliver hepatocyte cell therapy may enable more efficient hepatocyte function.

The spleen, peritoneal cavity, and subcutaneous space have all been proposed as potential sites for artificial hepatocyte engraftment. Direct hepatocyte injection into the splenic pulp of rodents has been shown to support hepatic regeneration, although mainly through the translocation of cells to the liver via the splenic vein.60 The peritoneal cavity offers a large capacity and easy access; the encapsulation of hepatocytes with microcarriers or hydrogel-based hollow fibers is a potential alternative.61 The infusion of hepatocytes into the subcutaneous space, supported by sufficient matrix to facilitate engraftment and by factors to promote neovascularization, has been demonstrated in rodent models with evidence of limited hepatocyte function.62 However, because of the complex structure required to enable hepatocytes to perform their multiple functions effectively, it will be challenging for these extrahepatic sites to offer a clinically relevant solution.

The development of extracorporeal bioartificial liver support (BALS) systems, in which infused whole blood comes into contact with hepatocytes either directly or via a semipermeable membrane, may offer a solution. Potential indications for BALS mainly include short-term applications in acute liver failure (eg, after partial hepatectomy) and bridging therapy prior to liver transplantation. The requirements for intensive care unit administration and the high cost limit longer term applications. Porcine hepatocytes as well as allogenic cells have already been incorporated into BALS with some success and have improved physiological parameters.63 A survival benefit in fulminant/subfulminant hepatic failure has been reported.64 Allogenic hepatocytes are in extremely short supply, with porcine hepatocytes potentially carrying the risks of anaphylaxis and infection (eg, porcine endogenous retrovirus).65 Ex vivo generated hepatocytes would avoid these risks.

A range of potential hepatocyte therapy delivery mechanisms are currently being explored. Because of the recent clinical application of BALS systems, the incorporation of ex vivo differentiated hepatocytes into such systems may seem the most likely mechanism for the delivery of hepatocyte stem cell therapies in the near future.


Bone marrow stem cells (BMSCs) have long been recognized as possessing the potential to support hepatic regeneration since the discovery of donor-derived cells in the livers of patients who had undergone bone marrow transplantation.66, 67 Subsequently, a distinct mobilization of bone marrow–derived cells has been observed after strong hepatic regenerative demand stimulated by, for example, partial hepatectomy or liver transplantation.22, 23, 68

Bone marrow contains HSCs, which are responsible for renewing circulating blood elements, and MSCs, which contribute to a wide range of mesenchymal tissues. Hypotheses about the mechanism by which BMSCs may contribute to liver regeneration have included transdifferentiation into hepatocytes, cell fusion creating hepatocyte cell hybrids, and paracrine effects.29, 69, 70 Although transdifferentiation of HSCs into hepatocytes has been reported, such events have not been found to occur at a clinically relevant magnitude.28-30, 71 HSCs have been shown to rescue a model of metabolic liver disease; however, the model used (fumarylacetoacetate hydrolase deficiency) produces a very high selection pressure that may not be relevant or reproducible clinically.29 A subsequent analysis of this model demonstrated that the rescue was due to cell fusion of bone marrow–derived monocytes and diseased hepatocytes, with fused cells dividing to repopulate the liver.69, 72 In vitro differentiation of MSCs into a hepatocyte-like phenotype is possible, but how functional these cells are in vivo is debatable.73 However, BMSCs may exert beneficial effects through paracrine mechanisms and, in particular, by enhancing angiogenesis. Effective angiogenesis is crucial to the regenerating liver and sustains its rapidly increasing mass. The majority of BMSCs migrating to the liver commit to sinusoidal endothelial cells, which play a central role in coordinating angiogenesis.74-76

In rodent models, the benefits of BMSC mobilization or administration have been widely reported. Mobilization of HSCs results in reduced mortality in acute liver injury by facilitating vascular remodeling, and accelerates recovery in chronic liver injury through the induction of endogenous repair mechanisms.77, 78 The paracrine effects of bone marrow–derived MSCs can inhibit the proliferative and fibrogenic function of activated stellate cells and induce stellate cell apoptosis.70 Enhanced repopulation of necrotized liver tissue by endogenous cells, protection against oxidative insults, selective leukocyte emigration with reduced fibrosis, and up-regulation of hepatocyte survival signals have all been attributed to the paracrine effects of bone marrow–derived MSCs.79-82 The injury-recovery interaction is certainly complex, however, with infused MSCs potentially contributing to liver fibrosis.52 Concerns over the phenotypic stability of engrafted MSCs require further evaluation in long-term preclinical transplant models.

Although there are concerns about the phenotypic stability of MSCs, the clinical application of HSCs in the treatment of hematological malignancy has been ongoing for many years. As such, the behavior and safety profile after human administration of HSCs is well established. Certainly, the focus of translational studies has turned to HSCs, which have emerged as the most promising cell type in this area.

Delivering Bone Marrow–Derived Cell Therapies

Obtaining BMSCs centers on either direct bone marrow aspiration or HSC mobilization with granulocyte colony stimulating factor (G-CSF) and collection from peripheral blood. Concerns have been raised regarding the use of G-CSF and specifically the potential for G-CSF administration to promote tumor growth and the risk of spontaneous splenic rupture.83, 84 Although direct bone marrow aspiration can avoid these risks, it is itself associated with discomfort and other procedure-related complications.

After BMSC harvesting, flow cytometry and magnetic activated cell sorting can be used for HSC enrichment by identifying the classic HSC cell surface antigens CD34 and CD133.85 It must be noted that these cell surface markers do not represent homogeneous cell populations but rather represent a heterogeneous mix of immature hematopoietic and endothelial cells with a continually and reversibly changing phenotype depending on the state of activation.85, 86 Reliable cell identification therefore necessitates the development of alternative techniques, perhaps based on metabolic markers, that exploit the quiescent nature of stem cells through G0-related proteins or identify molecules implicated in stem cell trafficking.86-88

Administering BMSCs to optimize hepatic engraftment has focused on peripheral vein, portal vein, or hepatic artery infusion. Much experience has been gained in these techniques for other indications such as chemoembolization. Although the direct delivery of cell therapies to the injured liver would appear to be a superior method, caution must be exercised in using contrast agents to visualize the hepatic artery or portal vein in patients at risk of hepatorenal syndrome.89 In animal models, the onset of liver regeneration is first noticed in cells surrounding the portal vein of liver lobules12; peripherally infused stem cells have been shown to engraft in portal tract areas, and portal vein infusion has demonstrated high first-pass stem cell entrapment.90, 91 Portal vein infusion may therefore be proven to be an appropriate starting point. Risks of exacerbating portal hypertension should certainly be considered, however, with dose administration reflecting this. Peripheral infusion of BMSCs avoids many of the risks associated with portal vein or hepatic artery infusion and has been effective in reducing fibrosis and improving survival in rodent models of chronic liver failure.92, 93

Potential indications for BMSC therapies are diverse and range from supporting hepatocytes in the acutely injured liver to remodeling the cirrhotic liver. Hepatic function could be supported in patients with decompensated cirrhosis awaiting transplantation, and in the acute phase of injury, bone marrow–derived cell therapies could ameliorate the fibrotic response and protect resident hepatocytes.

Stem Cell Therapies and Clinical Trials

Autologous stem cells derived from bone marrow are the only stem cell type to have undergone clinical investigation to date. The autologous nature of this cell source offers advantages in terms of ethical considerations and risks of sensitization and cell rejection. Several phase 1 studies investigating the application of BMSCs to enhance hepatic regeneration have been conducted, and 4 trials currently recruiting patients with chronic hepatic disease are listed in the US National Institutes of Health database of randomized clinical trials.94 Published studies have been small and essentially have established the safety and feasibility of the direct hepatic administration of BMSCs, rather than demonstrating efficacy, although benefits have been reported (Table 2). Diverse methods of BMSC isolation and selection, ranging from highly selective methods to less discriminatory ones, reflect current uncertainties and difficulties in selecting the appropriate cell subtype.

Table 2. Autologous Stem Cell Therapies and Clinical Studies in Liver Disease
Author (Year)IndicationCell SourceType of Stem CellDelivery of Stem CellsTreatment Group (n)Control Group (n)Primary OutcomeFollow-Up (Months)Improvement Shown?
  • *

    This includes am Esch et al.7 (2005).

  • This includes Gordon et al.9 (2006).

Yannaki et al.95 (2006)Chronic liver diseaseG-CSF–mobilized peripheral bloodMononuclear cells, enriched CD34+Peripheral vein20Child-Pugh/MELD12Yes
Terai et al.11 (2006)Chronic liver diseaseBone marrow from iliac crestCD34+/CD45/C-kitPeripheral vein90Child-Pugh6Yes
Fürst et al.8 (2007)*Enhanced liver volume prior to liver resectionG-CSF–mobilized peripheral bloodCD34+Portal vein67Daily liver growth rateN/A (preoperative)Yes
Mohamadnejad et al.89 (2007)Chronic liver diseaseBone marrow from iliac crestCD34+Hepatic artery40MELD6No
Mohamadnejad et al.99 (2007)Chronic liver diseaseBone marrow from iliac crestMSCsPeripheral vein infusion40MELD12Yes
Lyra et al.10 (2007)Chronic liver diseaseBone marrow from iliac crestMononuclear enriched cellsHepatic artery100Serum bilirubin, albumin, and INR4Yes
Levicar et al.96 (2008)Chronic liver diseaseG-CSF–mobilized peripheral bloodCD34+Hepatic artery (n = 2) or portal vein (n = 3)50Serum bilirubin and albumin6-18Yes
Pai et al.97 (2008)Chronic liver diseaseG-CSF–mobilized peripheral bloodCD34+Hepatic artery90Child-Pugh3Yes
Khan et al.98 (2008)Chronic liver diseaseG-CSF–mobilized peripheral bloodCD34+Hepatic artery40Serum bilirubin and albumin6Yes

Cohorts have included patients with chronic liver failure (n = 47) or patients with inadequate future liver remnant volume prior to partial hepatectomy (n = 6). The etiologies of chronic liver failure across the study groups have been diverse and have included alcohol alone (n = 15), hepatitis C virus alone (n = 10), cryptogenic cirrhosis (n = 8), hepatitis B virus alone (n = 7), alcohol plus hepatitis C virus (n = 2), alcohol plus hepatitis B virus (n = 1), autoimmune hepatitis (n = 2), primary biliary cirrhosis (n = 1), and chronic cholestatic liver disease (n = 1). The majority of studies have attempted to enrich harvested BMSCs for HSCs or MSCs, whereas others have simply concentrated unsorted mononuclear cells prior to infusion. A range of biochemical and clinical parameters, including the Model for End-Stage Liver Disease (MELD) score, Child-Pugh score (a prognostic score for chronic liver failure), liver volumetry, and liver function tests, have been assessed, with the majority of studies suggesting modest clinical benefit.

Following research in rodent models suggesting that BMSCs may lodge in the cirrhotic liver after massive bone marrow mobilization,100, 101 Gaia et al.102 and Spahr et al.103 administered G-CSF to small cohorts with chronic liver disease (n = 8 and n = 13). Clinical improvement was noted in some (n = 4), and it could be attributed to the mobilized cells or perhaps the direct effect of G-CSF itself on endogenous hepatic repair mechanisms.77

Because of the publication bias associated with any new technique, the results from these small phase 1 studies should be interpreted with caution. Conducting blinded, randomized controlled trials may seem ethically questionable at present because of the invasive procedures required for BMSC isolation and administration. Randomized trials may ultimately be required to establish genuine efficacy, although recruiting sufficient numbers of patients to adequately power studies may prove difficult with the great heterogeneity of suitable patient populations. Initial evidence supports the feasibility of BMSC administration in hepatic disease, but the range of protocols used raises numerous questions about the appropriate delivery of stem cell therapy in hepatic disease. Indeed, determining the fate of infused cells remains a major barrier to understanding the potential of stem cell therapy and necessitates imaging modalities capable of monitoring cell distribution.

The current inability to track transplanted or infused cells in human subjects represents a major challenge in further developing and understanding stem cell therapies. Currently, magnetic resonance imaging with supermagnetic iron oxide–labeled cells appears to offer the best cell-tracking opportunity. However, it is limited by dilution of the marker with cell division and the potential for transfer to other cells.104 Positron emission tomography may have a future role, although currently genetic modification is required to label cells, and this risks the alteration of cell properties.104

In conclusion, stem cell therapies in hepatic disease can be segregated into those that contribute directly to the hepatocyte population and those that may perform supportive roles. BALS systems using ex vivo differentiated hepatocytes and the supportive function of bone marrow–derived stem cells have emerged as the most promising mechanisms of cell therapy delivery to date. The type and severity of liver insult will likely affect the efficacy of stem cell interventions, with advanced hepatic cirrhosis perhaps the most challenging target. The ability to track stem cells in vivo and accurately monitor function will be crucial to understanding their role in future hepatic therapeutics. Ultimately, the most appropriate hepatic disease targets for stem cell therapies will become apparent as mechanisms of stem involvement in hepatic regeneration are further elucidated.