FormalPara Key Points

Inflammatory bowel disease is not adequately treated with conventional therapy.

Mesenchymal stem cell therapy has proven safe and effective in perianal fistulizing Crohn’s disease.

Mesenchymal stem cell and/or Treg therapy may be effective for luminal disease.

1 Introduction

Inflammatory bowel disease (IBD) is an idiopathic chronic transluminal inflammatory disorder of the gastrointestinal tract that is increasing in prevalence for unknown reasons [1, 2]. The Federal Drug Administration approved infliximab for the treatment of moderate to severe Crohn’s disease (CD) in 1998 [1]. This revolutionized the medical management of IBD from the use of glucocorticoids and immunosuppressants to monoclonal antibodies, or biologics, several more of which were subsequently licensed for use in IBD. Concurrently, greater use of a top-down or treat-to-target approach evolved, with the goal current medical therapy being not only symptom control but also mucosal healing as evaluated by endoscopy. Unfortunately, while biologics and, more recently, novel small molecule inhibitors, have improved disease control in many patients, at least one third of patients have primary nonresponse [2,3,4,5,6] and another third of patients experience secondary loss of response [7, 8]. Thus, despite several advanced therapies, and the increased morbidity and costs associated with cycling through these therapies, up to 60–80% with CD still require an intestinal resection [9, 10], and 20% with ulcerative colitis (UC) will undergo a proctocolectomy for medically refractory disease [11]. There is, therefore, an ongoing need for novel therapeutics. Cell-based therapies have recently been increasingly evaluated for the treatment of IBD and its associated phenotypes as one such novel class of potential therapeutics. The aim of this article is to walk through the current literature regarding cellular therapeutics for IBD, their limitations, and potential future technological advances to overcome these limitations.

2 Perianal Fistulizing Disease

Crohn’s disease (CD), a chronic transmural inflammatory disease of the gastrointestinal tract, continues to increase in incidence for unknown reasons [12]. According to population-based studies, at least 26% of patients with CD will develop perianal fistulas in the first two decades following diagnosis [13,14,15,16] particularly those with colonic and rectal involvement [13]. Up to 10% of women will develop a rectovaginal fistula [17, 18], another particularly devastating phenotype of perianal Crohn’s disease, which has a significant impact on quality of life due to the open communication from the rectum/anal canal to the vaginal wall [19]. Patients with rectovaginal fistula complain of stool and air per vagina, and some women will also report purulent discharge, dyspareunia, perineal pain and tenderness, and recurrent urinary tract infections [17, 20]. Perianal fistulizing patients experience significant morbidity due to pain, persistent drainage, recurrent perianal sepsis, and ongoing need to access medical care resulting in increased costs [21, 22] and impaired quality of life [22].

Unfortunately, perianal fistulizing Crohn’s disease is notoriously difficult to treat with 37% of patients experiencing refractory disease [23]. The most well studied agent for the treatment of fistulizing CD is infliximab, but less than one third of patients achieve fistula healing with this drug alone [19, 24]. In addition, biologic therapies have potential side effects including an increased risk of opportunistic infection [25]. Unfortunately, patients often cycle through several medications without benefit, all of which have potentially significant side effects. The lack of effectiveness means that > 90% undergo multiple surgical interventions [16] putting them at risk of incontinence [26]. The large number of surgical options including fistulotomy, plug and glue, anal/rectal flaps, or tissue interposition grafts, underscores that no one intervention is particularly effective. In addition, a normal rectum and anal canal are required for local surgical intervention, which is often not the situation in Crohn’s disease, greatly limiting the number of eligible patients for traditional surgical intervention.

Those with inflammation may have a seton placed (Fig. 1) in conjunction with initiation of infliximab, which can result in healing in up to 56% of patients; many, however, live with setons their entire life [27]. When surgical intervention is performed, long term success is limited by high rates of recurrence ranging between 25 and 50% [17, 28]. Therefore, up to 40% of patients will end up with a permanent diversion or a proctectomy, some with chronic non-healing perineal wounds [29, 30]. These results have spurred significant interest in identifying better treatment options that have the potential for improved efficacy without a risk of incontinence.

Fig. 1
figure 1

Seton placed into a perianal fistula tract

Full size image

3 Utilizing a Novel Therapeutic

The goal of a novel therapeutic to treat inflammatory bowel disease if first safety and secondarily efficacy as measured by disease response and patient quality of life [31]. Consequently, designing clinical trials that capture safety, efficacy, and quality of life (QOL) are key to positioning a novel therapeutic in the future. For fistulizing disease, symptom improvement along with improved QOL may be considered a success for some, while others are looking for complete fistula closure. Given that fistula patients often have a luminal disease component, patients should be kept on concurrent medical therapy to optimize luminal disease control while surgically and locally addressing the fistula tract with a novel cell therapy. This is in contrast to the luminal disease studies in which patients undergo a washout of their medical therapy to understand the potential benefit of a novel cell therapy on their luminal disease control.

3.1 Mesenchymal Stem Cells for Fistulizing Crohn’s Disease

Mesenchymal stem cells (MSCs) exist in almost all tissues [32,33,34], and are believed to reduce inflammation due to their intrinsic immunomodulatory properties. Recently, success of MSCs in treating severe inflammatory disorders, such as graft-versus-host disease (GvHD) [35], systemic lupus erythematosus [36], myocardial infarction [37], multiple sclerosis [38], and Crohn’s disease (CD) [39] has highlighted the therapeutic benefit of MSCs [40, 41]. In addition to being multipotent, MSCs act as immunomodulators to regulate T-cell activity and promote tissue repair [42]. Their immunomodulatory properties are thought to be carried out by migration to sites of active inflammation or tissue injury, secretion of anti-inflammatory molecules, such as interleukin-10 (IL-10), HGF, TGFβ1 [43], and indoleamine 2,3-dioxygenase (IDO) [44], and paracrine signaling to nearby cells to maintain the local anti-inflammatory environment [45, 46]. By influencing cytokine secretion profiles, MSCs can modulate the function of various immune cell types including lymphocytes, dendritic cells, and macrophages. Therefore, MSCs’ ability to change the local inflammatory cell milieu, migrate to sites of inflammation, and dampen immune responses underscores the escalating interest in using MSCs to treat IBD.

Locally administered MSCs are a promising option for complex perianal fistulizing CD (Fig. 2); even the American Society of Colon and Rectal Surgeons have updated their national guidelines to reflect the current literature to support MSC utilization [47]. MSCs can be harvested from bone marrow or from adipose tissue. Depending on the medical need, they can be collected from the patient’s own body (autologous) or from another donor (allogenic). The first successful use of locally administered MSCs for the treatment of a refractory rectovaginal fistula in the setting of CD was reported in 2003 [48]. These promising results generated a wave of phase 1, phase 2, and phase 3 trials to study the safety and efficacy of using MSCs to treat perianal CD (Table 1). Despite the heterogeneity in protocols using allogeneic or autologous MSCs, derived from both bone marrow or adipose tissue, administered at various doses, delivered as a singular or repeat injection, and delivered with or without scaffolding, the results of all completed trials have been very encouraging with regard to both safety and efficacy [42, 49]. Safety has focused on the rate of perianal abscess formation and perianal pain to determine whether injection of MSCs increases either. Efficacy has been largely defined as epithelization of the external opening, lack of drainage to palpation of the fistula tract at the external opening and lack of a fluid collection greater than 2 cm × 2 cm × 3 cm in any two of three dimensions. Given there are no approved cell-derived therapeutics to date with the Food and Drug Administration (FDA) for perianal fistula, these endpoints have not yet been provided in FDA guidance documents, but have been consistently used across the clinical trials performed to date, and will likely be used for drug approval by regulatory authorities in the future.

Fig. 2
figure 2

Injection of mesenchymal stem cells into a perianal fistula tract

Full size image
Table 1 Clinical trials of mesenchymal stem cells for perianal fistulizing Crohn’s disease
Full size table

With regard to safety, there have been no adverse or serious adverse events related to investigational product. Rather, the most common adverse events are pain at the injection site and perianal abscess formation which have been equivalent in the treatment and control groups. With regard to efficacy, data from a randomized placebo-controlled phase 3 clinical trial, ADMIRE, demonstrated significant efficacy as compared with placebo in the MSC treated group at six months (51 versus 36%; p = 0.021) and one year (56 versus 39%; p = 0.010) [50, 51].

Interestingly the durability of MSC healing has also been demonstrated by the long-term follow-up of ADMIRE clinical trial patients. Healing in the treatment versus placebo group was significantly improved at 6 months and this was maintained at 1 year, respectively [50, 51]. Even longer duration of follow-up found that the proportion of patients in ADMIRE with healed fistulas at 52, 104, and 156 weeks was 67, 54, and 54%, respectively, again underscored the durability of response over a longer follow up period [52]. Additionally, although a much smaller cohort, a group from Belgium showed that healing from a single dose of MSCs in Crohn’s perianal fistula was sustained to four years [53]

Given the promising healing rates in perianal disease in phase 3 clinical trials, additional phase 1B/2A clinical trials of approximately 20 patients have been performed showing safety and efficacy in complex perianal, rectovaginal, and pouch fistula phenotypes [54, 55]. This greatly expands the field’s experience with MSCs in multiple phenotypes of complex fistulizing disease, providing further data to support cell therapy in any phenotype of a Crohn’s related fistula. In addition, safety data has been reported in pediatric patients with Crohn’s-related perianal fistulizing disease [56].

3.2 Mesenchymal Stem Cells for Luminal Inflammatory Bowel Disease

Several animal studies have been conducted to demonstrate reversal of colitis with mesenchymal stem cells utilizing several modes of delivery including intraperitoneal injection [57, 58], direct rectal wall injection [59], intravenous delivery [59,60,61,62], and intra-arterial delivery. One study showed superiority of intravenous versus intraperitoneal delivery. However, again, due to the intravenous route resulting in pulmonary trapping, which can cause adverse effects such as embolus formation and post-transplantation pneumonitis, studies have focused on the safety and efficacy of intra-arterial delivery [63]. Thus, additional animal studies have now been performed demonstrating the safety of intra-arterial injection of MSCs specifically for the treatment of dextran sodium sulfate (DSS)-induced colitis. A group from Sweden injected 5 × 106 111In-oxine-labeled human decidual stromal cells (DSCs) into four rabbits via the superior mesenteric artery (SMA) using an angiographical system. The group found that selective injection of human DSCs to the rabbit SMA did not result in acute embolic complications or pulmonary trapping. Rather, selective injections to the SMA resulted in uptake distributed in the intestine supplied by the SMA and in the liver, indicating that this approach could significantly increase the fraction of injected DSCs reaching the target tissue without adverse events. Intravenous delivery corroborated pulmonary trapping. Another study looked at a study of mesenteric injection of adipose derived mesenchymal stem cells relieved colitis in rats without any safety concerns [64]. Another study looked at intra-arterial versus intra-venous delivery of MSCs intravenously versus intra-arterially and found that intra-arterial delivery was safe and had less uptake in the lungs [65].

Despite the large number of animal studies performed, only a handful of human studies have assessed systemic delivery of MSCs for IBD [66,67,68]. A study in patients with CD used Remestemcel-L, an allogeneic bone marrow derived MSC product (NCT00294112), which randomized patients to receive two infusions 1 week apart of either 2 million cells or 8 million cells/kg, intravenously. There were no reported infusion reactions, five mild adverse events unrelated to investigational product, and one serious adverse event (anemia) that were felt to be unrelated to investigational product. This phase II trial spurred a phase III clinical trial (Osiris protocol 610 NCT01233960) of four infusions over 2 weeks of either 600 million cells, 1200 million cells or placebo for medically refractory CD. The trial was stopped due to problems with protocol design and desire for open-label therapy, but there were no safety concerns. The only study of systemic MSCs for UC was a phase I/II trial of 70 UC patients treated by delivery into a peripheral vein or into the superior mesenteric artery versus placebo. Clinical response and remission were increased in the treatment group as compared to the control group, and there were no adverse reactions after MSC delivery [68]

Thus, overall, the intravenous deliver y of MSCs for medically refractory IBD has not significantly progressed. However, anecdotally, when we were treating patients in our own clinical trials of MSCs for the treatment of perianal Crohn’s fistulizing disease (ClinicalTrials.gov identifier: NCT04519671), we noted that proctitis would also improve after the direct injection near the internal opening of the fistula tract on the anal/rectal side. While previous clinical trials of the systemic delivery of MSCs for both luminal CD and UC have proven largely ineffective compared to conventional treatment approaches [66,67,68], none have used a method of direct targeted local delivery of cells, as has been utilized in clinical trials for perianal CD. Thus, because it has been well established that there is significant pulmonary trapping following intravenous delivery of MSCs [69], and there has been resolution of proctitis with direct injection, we sought to investigate the efficacy of MSCs delivered locally via endoscopic delivery, as is done in the setting of perianal disease, to treat the local site of inflammation directly. We performed two phase 1B/2A clinical trials of endoscopic delivery of MSCs in medically refractory ulcerative colitis and Crohn’s disease, both of which underscored the safety profile of MSC delivery and early limited efficacy [70, 71]. Future trials should be done to understand optimal dosing and dosing frequency with targeted efficacy outcomes.

3.3 Limitations of Mesenchymal Stem Cell Therapy

MSC properties vary depending on their tissue source within a patient, patient themselves, and culture conditions with process variation, all of which introduces variability and uncertainty into the cell-based treatment regimen. Donor age and sex affect the function of MSCs, and cell properties and functions differ between MSCs isolated from adipose tissue, umbilical cord, bone marrow, or other sites. While adipose MSCs are easier to harvest through liposuction or en bloc biopsy, no clinical trial has directly compared adipose versus bone marrow MSCs with regard to safety and efficacy for perianal CD. Nor has a clinical trial compared safety and efficacy of autologous versus allogeneic MSCs. Important strengths and weaknesses have been well established for each. Autologous cells are self-derived and will not evoke antibody responses with further treatments. However, patients are subjected to additional procedures for cell harvest that are not without risk and potential morbidity. Additional time and expense are then incurred for harvesting, expanding and testing the MSCs for sterility and viability prior to patient delivery and there is emerging evidence regarding the variability in MSC function across donors. Specific to CD is that MSCs from patients with CD do not appear as anti-inflammatory or immunomodulatory based on in vitro studies as compared to healthy control patients [72]. In contrast to autologous MSCs, allogeneic cells provide an off-the-shelf product that can be administered in a point-of-care fashion rather than requiring additional steps to harvest and expand the cells. In addition, since allogeneic MSCs are immunoprivileged with low expression levels of HLA class I and lack class II or costimulatory molecules, allogeneic MSCs may not generate antibodies in patients that would hinder efficacy with subsequent use. Nearly three times as many patients have been treated with allogeneic as compared to autologous MSCs. Similarly, nearly ten times as many patients have been treated with adipose derived MSC as compared with bone marrow MSC, likely due to the ease of adipose tissue harvest as compared with bone marrow harvest suggesting that ease of use is an important factor in MSC technology uptake.

Darvadstrocel (AlofiselTM), an allogenic adipose-derived viable MSC therapy, was approved in the European Union in 2018 and Japan in 2021 for the treatment of refractory complex perianal fistulizing CD in adults. Currently, it is the only MSC product approved for use by a regulatory body for the treatment of any phenotype of IBD. The US Food and Drug Administration (FDA) granted regenerative medicine advanced therapy (RMAT) designation to darvadstrocel in May of 2019 for complex perianal fistulas in adults with CD and its approval is currently under review as enrolment in the ADMIRE II phase III clinical trial was recently completed. The challenge with darvadstrocel is justifying the high cost of therapy when the placebo rate in the phase 3 clinical trial and subsequent follow up were so high; the delta in placebo and treatment was only 15%. This high placebo rate calls into question the actual utility of the MSCs, especially when the cost can be prohibitively high at £54,000 [50, 51].

In addition, while MSCs are promising regarding safety profile and clinical efficacy, the infrastructure required results in limited penetration to practice following approval. The challenge with viable MSCs is the infrastructure needed to deliver cell therapy and the limited number of sites with that capability. MSCs have a short shelf life of cell viability—approximately 48 h from the laboratory to the operating room—and are typically best used fresh, adding limiting complexity to procedures [31]. In addition, scalability and reproducibility is a challenge. The potential for tumorigenesis and teratoma formation is genuine. Lastly, there is a significant concern for alloimmunity. Timing of cell shipment, cost of shipment and product, limited starting material, and need for cell therapy pharmacies translates to limited clinical use occurring only in centers that have the required infrastructure and patients or healthcare systems that can afford the product. Thus, a secretome product offers the potential benefit of cell therapy while concurrently offering a scalable product, which could be offered in any clinical setting, regardless of clinical infrastructure.

3.4 Next Generation of Cell Therapy

There are well designed studies showing that paracrine effects mediated by the MSC secretome, including the extracellular vesicle (EV) compartments and noncompartmentalized biomolecules, are the most important aspect of MSC function related to local healing rather than transdifferentiation across cell types. In fact, one study using autologous bone marrow MSC found a significant increase in the percentage of circulating CD4+CD25 brightFoxP3+ regulatory T cells at 12 months (p < 0.01) and mucosal FoxP3+ regulatory T cells in the inflamed areas (at 12 months) (p < 0.001), suggesting that bone marrow MSCs are able to recruit relevant immune cells to local environment of the fistula. Therefore, the successes achieved with the use of viable MSC preparations to treat fistulas associated with perianal CD, as summarized above, suggest that a preparation of MSC-derived EVs may also be effective, safe, and considerably more convenient.

Preclinical studies have confirmed that EV therapy may be effective in the treatment of IBD. In the DSS-induced mouse model of colitis, human umbilical cord derived exosomes injected intravenously homed directly to the diseased colons within 12 h, and macrophage infiltration and disease severity were significantly reduced [73]. Anti-inflammatory IL-10 expression was reduced while expression of proinflammatory TNF-α, IL-1β, IL-6, iNOS, and IL-7 was decreased in the colons of these DSS-induced diseased mice. In another relevant study, human dental pulp MSCs were expanded, induced to overexpress HIF-1α, and primed with proinflammatory stimuli before harvesting the “boosted EVs” from the cultures [74]. The EVs were then shown to immunomodulate macrophages from an M1 proinflammatory phenotype to an immunosuppressive M2 phenotype in vitro, as well as following intraperitoneal injection, to increase the infiltrating M2:M1 ratio and attenuate disease status in a mouse TNB induced model of colitis. EVs derived from human placenta MSCs were evaluated in a recent phase I study wherein 5 of 11 patients with fistulizing disease showed complete tract resolution over 6 months; only one failed to show any improvement and no safety events were reported [75]. These studies are direct in vivo demonstrations of the potential for MSC EVs to favorably modify the inflammatory basis of colitis and contribute to disease resolution and support our hypothesis that direct injection of the IMP at the diseased area will promote resolution of inflammation-driven perianal fistulas associated with Crohn’s disease.

Given these preclinical studies, and potential with a MSC–EV-derived therapy, there are now clinical trials underway to study ExoFloTM, a MSC-derived EV product for the study of luminal medically refractory Crohn’s disease (IND27959; NCT05130983) via intravenous delivery, luminal medically refractory ulcerative colitis (IND27959; NCT05176366) via intravenous delivery, and refractory perianal Crohn’s disease (IND29323; NCT05836883) via direct injection. The results of these early clinical trial efforts will provide additional insight as to the next generation of cell therapy with EV therapy, providing an off the shelf, truly scalable regenerative medicine approach.

4 Beyond Mesenchymal Stem Cell Therapy—Regulatory T Cells

Regulatory T cells (Tregs) are nonredundant regulators of intestinal tolerance [76]. Quantitative and qualitative defects in mucosal Tregs have been identified in patients with Crohn’s disease leading to the hypothesis that correction of these defects may be a therapeutically tractable way to reset a disordered homeostatic immune response in the gut [77]. Over the past few years, modulation/enhancement of endogenous Tregs with low-dose IL-2 or an engineered derivative have been largely clinically ineffective. Therefore, attention has increasingly started to turn to the adoptive transfer of autologous Tregs, expanded in vitro and modified to traffic to the intestine. The most clinically advanced of these programs in refractory Crohn’s disease is the TRIBUTE study (NCT03185000), which involves the transfer of polyclonal Tregs that have been induced to express high levels of the gut homing integrin α4b7. The efficacy of this approach, either alone or in combination with MSCs/EVs in refractory or fistulizing disease remains to be determined.

5 Conclusions

Cell therapy offers a new frontier of medical therapy to treat inflammatory bowel disease. Much of the evidence to date comes from the use of cell therapy in perianal fistulizing Crohn’s disease. Fortunately, the studies to date have underscored the safety of cell therapy, and initial work has highlighted the potential efficacy. However, there are still several unanswered questions such as the mechanism of action of cell therapy on IBD and how to scale cell therapy to treat thousands of patients with IBD. This will need to be better illustrated in future studies.