Stem cells as potential novel therapeutic strategy for inflammatory bowel disease

Abstract

Hematopoietic stem cell transplantation and mesenchymal stromal cell therapy are currently under investigation as novel therapies for inflammatory bowel diseases. Hematopoietic stem cells (HSC) are thought to repopulate the immune system and reset the immunological response to luminal antigens. Mesenchymal stromal cells (MSC) are cells that have the capacity to differentiate into wide variety of distinct cell lineages and suppress immune responses in vitro and in vivo. Recent results from animal models and early human experience in graft-versus-host disease but also Crohn's Disease suggest that ex vivo expanded MSCs may have clinically useful immunomodulatory effects.

Hematopoietic stem cells, Mesenchymal stromal cells, Transplantation, Crohn's Disease

1 Introduction

Despite the improvement in medical therapy of inflammatory bowel disease (IBD) with the introduction of anti-TNFα compounds, disease control remains hard to achieve in many patients. Adult stem cells are currently under investigation for a variety of inflammatory disorders. This article reviews the potential of stem cells therapy for IBD. We will summarize reports on hematopoietic stem cell transplantation (HSCT) in inflammatory bowel disease and we will introduce the possible advantages of mesenchymal stromal cell (MSC) therapy.

1.1 What is a stem cell?

A classical definition of a stem cell is a cell that has the capacity for self-renewal and the ability to give rise to one or more types of differentiated progeny.¹,² Self-renewal is defined as the ability of a cell to proliferate while it maintains its proliferation and differentiation potential. Stem cells are known to exist in different tissues but their frequency, exact role and identity are generally not well understood. Both in animal models and in patients, it appears that bone marrow derived cells play a role in the healing process following intestinal injury³ and that they may contribute to regeneration of various components mucosa.⁴–⁶ The bone marrow contains at least two types of stem cells. One population consists of CD34 positive hematopoietic stem cells committed to differentiate into all blood cell types, including the myeloid and lymphoid lineages. A second population of stem cells remains less well characterized. These non-hematopoietic stem cells are thought to support hematopoiesis and are variously known as mesenchymal stem cells, marrow stromal cells and more recently, mesenchymal stromal cells, all designated by the acronym MSC.⁷

2 Hematopoietic stem cell transplantation

Although more than 50 years ago HSCT was introduced as a treatment for injury, it is now principally used for hematologic and lymphoid cancers.⁸ Evidence that HSCT is an effective treatment for autoimmune diseases come from animal models⁹,¹⁰ and case reports from HSCT recipients with coexistent autoimmune disease.¹¹ Ever since, autologous HSCT has been performed in more than 700 patients with autoimmune disease.¹² The most frequent indications being systemic sclerosis, multiple sclerosis, rheumatoid arthritis and systemic lupus erythematosus.

HSCT includes conditioning (high dose chemotherapy, total body irradiation and/or anti-lymphocyte antibodies) which completely eliminates bone marrow cells of the host followed by the infusion of either autologous or allogeneic stem cells. The HSCs are either directly harvested from the marrow or mobilized from marrow or blood and harvested by apheresis. High dose immune ablation is an intensive treatment with risks of severe complications which on rare occasions have been fatal. In autologous transplantation, the individual's own hematopoietic stem cells are harvested to be returned after conditioning. The graft is typically depleted of T cells to avoid the reinfusion of autoreactive T cells. In allogeneic transplantation, the HSCs come from a donor, usually a HLA matched sibling. In addition to the complications associated with conditioning, allogeneic HSCT is associated with a much higher transplant-related morbidity, such as graft-versus-host disease, aplastic anemia, and hematological malignancies, and also a higher mortality (15–25% vs. 3–5%¹³) due to a considerable risk on graft-versus-host disease. In view of the risks related to allogeneic transplantation most patients treated for autoimmune disease have received autologous transplants.

2.1 Hematopoietic stem cell transplantation in IBD patients for other indications

2.1.1 Autologous HSCT

The possibility that autologous HSCT could be an effective treatment for IBD was suggested by the improvement of the clinical course of disease in patients with CD that received autologous transplantation for other indications.¹⁴–¹⁸ The first published abstract dates from 1993 and described two year clinical remission in two patients with active IBD treated with autologous HSCT for breast cancer.¹⁴ Long term clinical remission after autologous HSCT was reported in a patient with clinical disease control of CD up to 7 years following transplantation for non-Hodgkin lymphoma.¹⁶ A similar result was obtained in a 30 year old patient with a ten year history of severe CD who developed Hodgkin's disease and remained in complete treatment-free remission of both diseases 3 years after autologous HSCT¹⁹ and in a patient who had normal findings during ileo-colonoscopy at 1, 2, 3 and 5 years after transplantation for acute myeloid leukemia.¹⁸

2.1.2 Allogeneic HSCT

Similar to the first experience with autologous HSCT in IBD, the effect of allogeneic HSCT on IBD was initially described in patients treated for hematological malignancies. The first case report was published in 1998 and described a 35 year old male free of symptoms and signs of CD 8 years post allogeneic marrow transplantation for acute leukemia.²⁰ A second report in the same year described 6 patients that underwent allogeneic transplantation for leukemia between 1962 and 1982.²¹ In this report, 5 out 6 patients had active CD at time of transplantation. One patient died of septic complications 97 days after transplantation; the other 5 patients remained free of disease activity for more than 1 year post-transplantation. Only one out of these five patients relapsed during follow up with a duration of up to 15 years post transplantation. Interestingly, the only patient that developed a mixed donor-host hematopoietic chimerism following allogeneic HSCT continued to have active CD before death by suicide. In a retrospective study by Ditschkowski et al., 10 out of 11 patients remained free of symptoms following allogeneic HSCT for hematological malignancy with a median follow-up time of 34 months.²² In a case report published recently,²³ a 41 year old man underwent allogeneic HSCT for lymphoma. Following transplantation his bowel symptoms ceased and he was able to stop all immunosuppressive drugs. Eighteen months after transplantation colonoscopy showed no evidence of Crohn's disease activity. Remission of ulcerative colitis following allogeneic HSCT has also been described.²⁴ Two patients, each with a long history of psoriasis and ulcerative colitis, received an allogeneic HSCT for leukemia and remained in full remission 4²⁴ and 12 years²⁵ after transplantation.

The coincidental treatment of IBD with both autologous and allogeneic HSCT increased the interest in the possibility that stem cell transplantation could be of value in IBD. Autologous HSCs are infused only to shorten the post-HSCT neutropenic interval, in contrast with allogeneic HSCT in which the recipient's immune and hematologic system is replaced with that of a healthy donor without the genetic predisposition to Crohn's disease. In this light it has been proposed that the risk of disease recurrence may be higher after autologous HSCT.²⁶

2.2 Early studies on HSCT specifically for CD

The first reports of autologous HSCT specifically given for the treatment of Crohn's disease were published in 2003 and concerned five patients with severe disease activity refractory to conventional treatment and treatment with anti-TNFα antibody.²⁷–²⁹ No serious transplantation related complications were reported and all patients entered clinical remission but some of the colonoscopies showed persistent mild inflammation up to 1 year post-transplantation. A larger phase 1 study on twelve patients with chronic active, refractory CD also suggested that autologous HSCT can have a beneficial effect on CD activity. Besides fever, the autologous HSCT was well tolerated by the patients. After 15 months only one patient developed a recurrence of active CD. All others maintained a clinical and drug-free remission but similar to the patients described in the reports above with persisting nonsymptomatic histologic and/or radiologic evidence of CD.³⁰ Adverse effects included hematemesis from a Mallory–Weiss tear, a prolonged febrile course, C. difficle-induced diarrhea, and diarrhea after an upper respiratory infection.

Our own experience consists of three autologous HSCT in patients with refractory CD (Zelinkova et al., DDW 2005). Our protocol includes a mobilization phase with cyclophosphamide 4 g/m2 and granulocyte colony-stimulating factor (G-CSF) 2dd 5 μg/kg. Prior to transplantation, immune ablation is achieved using cyclophosphamide 50 mg/kg/day (4 days), anti-thymocyte globulin 30 mg/kg/day (3 days) and prednisolone 500 mg (3 days).

Autologous HSCT was offered to one patient suffering from extensive refractory small bowel disease, failing combination therapy of immunosuppressives and infliximab and for whom surgery was considered not to be an attractive alternative as it would most likely result in a short bowel syndrome. Although an initial clinical response after mobilization was observed, the patient experienced a clinical and endoscopic relapse within weeks. During the HSCT patient developed an allergic reaction to anti-thymocyte globulin that resolved with a standard treatment and the HSCT was completed successfully. In week 8 of the follow-up after HSCT, the patient was clinically in complete remission, gaining weight and without need for a supplementary parenteral nutrition. Endoscopically, a clear improvement of the inflammatory lesions was observed. As expected, residual strictures persisted. Two additional patients with refractory CD, unresponsive to anti-TNF therapy and experimental anti-CD3 (visilizumab) therapy, unwilling to undergo proctocolectomy, underwent successful (partial) autologous HSCT. One male patient with a complicated history of a refractory CD that failed infliximab and experimental anti-CD3 therapy underwent the mobilisation phase, had a partial response, and was subsequently transplanted. He regained full remission, both clinically and endoscopically. We restarted azathioprine therapy, and during a follow-up of 2 years the patient remains in remission. Our third patient is a female with a similar history, a completely refractory CD including failure of anti-TNF therapy. After the mobilisation phase she went in complete remission and, as a result, was not transplanted. She has now been in complete clinical and endoscopic remission for 2 years including an uncomplicated pregnancy.

This last patient is a good example of role of the immunosuppressive/cytoreductive effects of the conditioning regimen in the treatment success of HSCT. To evaluate the efficacy of HSC mobilisation followed by high dose immune ablation and autologous stem cell transplantation versus HSC mobilisation only, a multicenter, prospective, randomised phase III study has been initiated by the European Crohn's and Colitis Organisation (ECCO) in collaboration with the European Group for Blood and Marrow Transplantation (EBMT).³¹

2.3 Potential risks of hematopoietic stem cell transplantation

HSCT may be an effective treatment for CD but is also associated with a high morbidity and mortality.³² In 390 patients undergoing autologous HSCT for various autoimmune diseases a mobilization associated mortality of 1.5% and an overall procedure related mortality of 9% were found.³³ Early toxicity is related to direct organ damage either from the agents used or due to infection and bleeding during the 10–12 days of bone marrow aplasia following the immunosuppressive conditioning period. Late toxicity relates to malignancy development due to the chemotherapy and/or radiation exposure. In addition, HSCT is associated with complications such as veno-occlusive disease of the liver and acute and chronic graft-versus-host disease.³⁴ Although HSCT seems a reasonably successful treatment for CD it is clear that, given the considerable mortality rate of HSCT for autoimmune diseases, this treatment should only be considered to rare cases of CD. HSCT could be considered as a last resort for treatment in patients with debilitating disease refractory to all immunosuppressive drugs, including the different anti-TNFα compounds now available for treatment, and in patients in which surgery is not a treatment option.

3 Mesenchymal stromal cell transplantation

HSCT is thought to result in clinical remission in CD due to the combination of the immunosuppressive conditioning regimen and the replacement of the derailed lamina propria immune cells that maintain the disease. A novel emerging stem cell treatment may offer the benefit of immunosuppression without the need for conditioning chemotherapy, even when given as allogeneic transplant.

3.1 Mesenchymal stromal cells

Friedenstein and colleagues were the first to identify an adherent, fibroblast-like population of cells in bone marrow.³⁵ These cells are now known under various names but will be termed mesenchymal stromal cells or be referred to by acronym MSC throughout this review. MSCs turned out to have stem cell like properties. The first MSCs that have been described were bone marrow derived non-hematopoietic cells since they did not express hematopoietic cell lineage markers such as CD34, CD45 and CD14.³⁶,³⁷ Indeed, MSCs have now been isolated from a wide variety of tissues including fat, periosteum, hair follicles and scalp subcutaneous tissues, muscle, synovium, synovial fluid and periodontal ligament,³⁸–⁴⁷ placenta, umbilical cord blood, fetal blood, bone marrow, liver, lung and spleen⁴⁸–⁵⁰ suggesting that MSCs are not unique to the bone marrow environment but in stead widely distributed in vivo⁵¹. Depending on the specific culture conditions used, these cells have the potential to generate a wide variety of mesenchymal cell lineages, including bone,⁵² cartilage,⁵³ fat,⁵⁴ muscle,⁵⁵ tendon,⁵⁶ stroma⁵⁷ and cardiomyocytes.⁵⁸–⁶⁰ In vitro and in vivo studies have also indicated the capability of MSCs to differentiate into neural precursor,⁶¹,⁶² and hepatocyte⁶³ like cells. Very important for their potential clinical application is that MSCs are easily isolated as they adhere to plastic and are capable of substantial proliferation and expansion in culture.³⁷ Additionally, MSCs can be cryopreserved with no loss of phenotype or differentiation potential.⁶⁴,⁶⁵

4 Definition

In the absence of an agreed standardized marker, MSCs have been defined by a combination of phenotypic and functional characteristics. To further homogenize studies on MSCs the Mesenchymal and Tissue Stem Cell Committee of the International Society for Cellular Therapy (ISCT) has introduced three minimal criteria to identify human MSC⁶⁶ [Table 1. Three minimal criteria to identify MSCs as defined in the ISCT position statement].

4.1 Preclinical data on MSCs

Immunomodulatory functions of MSCs were examined in vitro by coculturing them with purified subpopulations of immune cells and report that MSCs suppress several functions of naïve and memory T cells,⁶⁷–⁷² B cells⁶⁷,⁷¹,⁷³ and natural killer cells⁷¹,⁷⁴,⁷⁵ as well as the differentiation, maturation and function of dendritic cells.⁷⁶–⁷⁸ Furthermore they have been shown to alter the cytokine secretion profile of the host immune cells to shift from a pro-inflammatory phenotype toward a more anti-inflammatory or tolerant phenotype in vitro.⁷⁹ Di Nicola et al. found that MSCs reversibly inhibited T cell activation by several different stimuli in vitro. In their experiments cell to cell contact was only partially responsible for these effects, in transwell experiments it was shown that the observed inhibition depended mainly on the production of soluble factors.⁸⁰

Animal studies indicate that, similar to their immunosuppressive capacities in vitro, MSCs also display immunosuppressive capacities in vivo. Murine MSCs have been used to successfully treat a T-cell mediated experimental model of multiple sclerosis in mice⁸¹ and administration of allogeneic MSCs led to prolonged skin graft survival when compared to control animals in immunocompetent baboons.⁷⁰ Systemically infused MSCs first show a wide-ranging distribution followed by homing to injured tissues⁵⁸,⁸² (including the gut⁸³–⁸⁵) where they may participate in tissue repair.⁸³,⁸⁵ In addition, following intravenous infusion MSCs can home to the bone marrow where they can persist in for an extended period of time.⁸⁴ However, only a small percentage of the infused MSCs are traceable, and the fate of the rest remains unknown.

Although thought to be mostly immune privileged, MSCs may under certain conditions also be subject to immune rejection. Data from a murine model suggest that, in nonmyeloablated hosts, allogeneic MSCs are able to mount a T-cell memory response and consequently are eliminated.⁸⁶ Comparable loss of immune privilege has been reported by others.⁸⁷

4.2 Clinical data on MSCs

Due to their immunosuppressive properties and their role in tissue repair MSCs seem a promising tool in immunoregulatory and regenerative cell therapy in a variety of human diseases. Furthermore, a major advantage of MSCs over HSCT is that infusion of MSCs is nonmyeloablative, thus without total body irradiation. The initial clinical trials were in patients with osteogenesis imperfecta,⁸⁸ followed by trials in which the immunosuppressive effects of the MSCs were used either to reduce the incidence of graft-versus-host disease after allogeneic HSCT⁸⁹ or as treatment of active disease, including GvHD of the gut refractory to conventional immunosuppressive therapy.⁹⁰ The effects of MSCs on GvHD suggested the potential use of MSCs in the treatment of other autoimmune disorders.

5 MSCs in Crohn's disease

Recently, the results have been published of a phase I clinical trial on cell therapy, in which autologous MSC transplantation was used for the treatment of fistulizing CD. In this study, nine fistulas in four patients were inoculated with autologous adipose tissue-derived MSCs. After 8 weeks 75 percent these fistulas were considered healed and no adverse effects were observed in any patient.⁹¹ A phase II trial on autologous adipose tissue-derived MSCs is underway.⁹²

In a press release, the American company Osiris Therapeutics, Inc. reported encouraging results of a phase II study in patients with moderate to severe CD unresponsive to standard pharmacologic interventions using Prochymal™, a type of MSCs derived from the bone marrow of normal healthy adult volunteer donors.⁹³ All patients were reported to show a significant reduction in disease activity by day 28 upon infusion of stem cells.⁹⁴ Unfortunately the trial has not yet been published in a peer reviewed journal for evaluation of these important claims made in the media. Osiris Therapeutics is reportedly enrolling patients for a phase III trial.⁹⁵

5.1 Potential risks of mesenchymal stromal cell transplantation

More than 100 patients have received MSCs and though acute toxicity appears low, little is known about long-term unwanted side effects.⁹⁶ Reports showed that MSCs stimulate the growth of cancers in mice⁹⁷,⁹⁸ and promote metastasis in mice.⁹⁹ Although an increased risk on tumor formation has never been confirmed in humans, patients should be thoroughly screened before MSC administration as the cells might enhance the growth of unknown cancer. Another concern is that extensively in vitro expanded stem cells may be prone to malignant transformation. In fact, human MSCs derived from adipose tissues have been shown to undergo spontaneous transformation after being passaged for a long time (4–5 months) in in vitro culture. However, no such abnormalities where reported by Bernardo et al. who proliferated bone marrow derived MSCs from 10 healthy donors for up to 44 weeks until reaching either senescence or passage 25.¹⁰⁰ Accordingly, it is of great importance to carefully characterize MSCs passaged in vitro both phenotypically and karyotypically to maximize safety for the recipient.

6 Conclusion

The clinical course of Crohn's disease can not be adequately controlled in a substantial group of patients despite treatment with an extensive repertoire of immunosuppressives, anti-TNF therapy and surgical interventions. Severe complications of the disease include peri-anal fistulizing disease.

Autologous HSCT has been evaluated for sever refractory cases. This treatment can be successfully used as a last resort in an attempt to control debilitating disease is associated with significant morbidity and mortality related to chemotherapy. The use of MSCs derived from either bone marrow or adipose tissue could be an alternative appraoch. This treatment does not work by replacement of the disease causing inflammatory cells in the lamina propria but is believed to involve active suppression of these immune cells by as of yet uncharacterized MSC derived signals at the site of inflammation. If effective, the big advantage of the use of MSCs is the fact that this treatment does not involve conditioning chemotherapy. MSCs are currently under active investigation as a treatment option for Crohn's Disease and encouraging preliminary data seem to support the further study of this new approach.

Tables

Table 1

Three minimal criteria to identify MSCs as defined in the ISCT position statement⁶⁶

MSCs must
1. Adhere to plastic when maintained in culture
2. Be able to differentiate into at least osteoblasts, adipocytes and chondroblasts under standard in vitro differentiating conditions (demonstrated by well established staining protocols)
3. Be positive (≥ 95%) for CD105, CD73 and CD90 and be negative (≤ 2% positive) for haematopoietic antigens like CD45, CD34 and markers for monocytes, macrophages (CD14 or CD11b) and B cells (CD79α or CD19). HLA-DR molecules should not be expressed on MSCs unless they are stimulated, e.g. by IFNγ.