Final Report Summary - MERLIN (MEsenchymal stem cells to Reduce Liver INflammation)
Executive Summary:
It is estimated that 29 million people in the EU have chronic liver disease, and it is the fifth most common cause of death. Most liver diseases involve inflammation that leads to liver damage. MERLIN is focused on developing a stromal cell therapy for the liver diseases primary sclerosing cholangitis (PSC) and autoimmune hepatitis (AIH). There are currently limited treatment options available for these conditions.
Researchers in MERLIN have looked at the effectiveness of Mesenchymal stromal cells (MSCs) against inflammatory liver disease in the laboratory, in order to inform and underpin a novel cell therapy. We have also explored mechanisms of action of MSCs and optimum conditions for MSC production.
Some of our key findings include:
• MSCs reduce markers of liver damage and inflammation in laboratory models of inflammatory liver disease, as well as the number of inflammatory cells in damaged areas.
• MSCs derived from both bone marrow (BM) and umbilical cord (UC) show positive effects.
• The route of infusion of UC MSC has no impact on its therapeutic effect (MSCs infused subcutaneously and intravenously were equally effective).
• The beneficial effect of MSCs on PSC is due, at least in part, to the extracellular vesicles (EVs) that MSCs secrete.
• The properties of MSCs can be influenced by the conditions under which the MSCs are cultured. Culturing protocols also merit due consideration.
• MSCs largely move to the lungs following administration, with an accumulation of dead MSCs in the liver 24h after infusion.
In MERLIN we used the results of our studies to design a clinical trial to test the safety of our MSC therapy and examine the effect of therapy on inflammation in patients with PSC and AIH. We also developed a process for MSC production from UC tissue that is consistent and aligned with regulatory requirements. NHSBT have manufactured specially selected MSCs (ORBCEL-C™, discovered by Orbsen Therapeutics) for delivery to patients in the trial. The MERLIN trial opened on 7 December 2018 and will continue (and be completed) after the project ends.
We believe that our work will set the stage for a new, MSC-based therapy for PSC and AIH, with potential application to other forms of liver disease and other inflammatory conditions. The project has delivered important data and lays the groundwork for the future tailored production and use of precisely characterised MSCs for particular applications.
In addition to potential benefits for individual patients and their families, improving treatments for liver disease will also benefit healthcare systems, through the reduction of costly liver transplant procedures and of care costs associated with chronic liver disease. From a health economics perspective, MSC therapies for immune and inflammatory diseases offer potential for significant savings in the longer term. Researchers in liver disease, immunology and regenerative medicine will benefit from the new knowledge generated by the project and the MERLIN trial. An additional benefit derived from the project will be the advancement of 3D imaging technology, which offers researchers new, better ways to carry out bio-distribution studies.
The MERLIN consortium is committed to building on the project in future collaborations, generating long-term research value and delivering societal and economic benefits through the development of advanced MSC therapies.
Project Context and Objectives:
The overall objective of MERLIN was to develop and validate (in a Phase 2a clinical trial) purified MSC as an entirely new treatment option for patients with inflammatory liver diseases PSC and AIH, in parallel with enhancing knowledge about the operation of MSCs.
Background
The prevalence of chronic liver disease has been estimated to be 6% (or 29 million people) in the EU (The European Liver Patients Association. 2005), with mortality rates estimated at 14.3 per 100.000 in the EU-25 in 2004, making it the 5th commonest cause of death (the EU Statistical Year Book, 2007). Most liver diseases have a significant inflammatory component that underpins liver damage & fibrogenesis, yet current therapies have limited effectiveness. Primary Sclerosing Cholangitis (PSC) is an example of chronic immune mediated liver injury, with a natural history of progression to liver failure. The incidence and prevalence rates for PSC vary from 0 to 1.3 per 100,000 inhabitants/year and 0 to 16.2 per 100,000 inhabitants, respectively. Liver transplant is ultimately the only viable treatment for many people with PSC.
Our challenge in MERLIN was to develop a therapy that alleviated the ongoing inflammation that drives fibrosis in liver injury. Using PSC and AIH as our paradigm reflected the inflammatory and immune mediated nature of these conditions, alongside the huge unmet need for therapy. We aimed to harness laboratory and clinical studies to effectively, and in a timely manner, answer our questions regarding optimised MSC therapy and how such therapy might be applied in the context of PSC and AIH.
MERLIN Key Objectives
As noted above, the main objective in MERLIN was to develop a Phase 2a clinical trial administering purified MSC as an entirely new treatment for PSC and AIH patients, in tandem with developing new knowledge about MSCs.
Some more specific and detailed objectives of the project described in the original programme of work are set out below.
Translational Aims (related to the MERLIN clinical trial)
The aims below were devised at the outset of the project to help us deliver the proposed clinical trial.
• To compare the safety and efficacy of ORB1+ and ORB1- MSC to PA-MSC in refined models of PSC – this aim was designed to help us select the most appropriate MSC-based product for use in the MERLIN trial.
• To ensure EU ATMP compliance and prepare an application for a Phase 2a Clinical Trial of ORB1 therapy in PSC – this aim involved the generation of an Investigational Medicinal Product Dossier (IMPD) Investigator Brochure (IB) and Clinical Trial Protocol for the MERLIN Clinical Trial Application to the Medicines Health and Regulatory Authority (MHRA) in the UK.
• To carry out a Phase 2a Clinical Trial of ORB1 therapy in PSC – this aim related to the validation of the selected MSC-based therapy in the MERLIN clinical trial.
• To establish the in vivo immunological footprint of MSC after infusion into pre-clinical models of PSC and also in clinical trial patients – this aim set out practical tests to be applied in order assess the impact of the therapy administered.
Development Aims (increasing knowledge about MSCs)
The following aims were identified at the start of the project in order to guide our studies into MSCs and help us develop a deeper understanding of MSC operation.
• To elucidate the mechanism of action of MSC in ameliorating liver inflammation -this aim required us to explore the role and mechanisms of action (MoA) by which MSC simultaneously alleviate liver inflammation and repair tissue damage.
• To improve the functional longevity of MSC – this aim required an assessment of culture protocols for MSCs, to determine at what point in the culture process their efficacy may be compromised.
• To reduce the immunogenicity of MSC – this aim focused on exploring how to modulate the potential susceptibility of MSC to immune cell recognition, in an effort to improve survival of MSCs post infusion.
• To improve the homing/localisation of MSC – this aim required us to consider the homing and biodistribution of MSCs post infusion and the most effective ways to administer MSCs.
• To generate the optimal MSC – this aim required the study of conditions which might optimise the operation of MSCs.
Legacy Aims (To maximise the long-term research and collaboration value of MERLIN).
Specific aims in relation to MERLIN’s long-term legacy were clearly articulated in the MERLIN programme of work:
• To agree a Memorandum of Understanding between the partners to guide long term collaboration with potential to expand and add new members.
• To agree standard operating protocols for isolation, culture, storage, transport and use of MSC.
• To publish training and best practice materials for the wider research community.
• To develop training opportunities across the consortium.
Conclusion
We put the aims and objectives outlined above into practical application as we embarked on the MERLIN work programme. We explored the Development Aims described and discovered new knowledge about MSCs. MERLIN partner BIO succeeded in further developing their 3D imaging technology (CryoVizTM) for high-throughput imaging and analysis. Different elements of our research emerged as more or less important as our studies generated novel results and directed our focus.
We applied the new information learned in the design of the MERLIN clinical trial, which is now underway. NHSBT have manufactured specially selected MSCs (ORBCEL-C™, discovered by Orbsen Therapeutics) for delivery to patients in the trial.
Details of the scientific work undertaken and key results achieved are set out in section 1.3 below.
Project Results:
Introduction
Mesenchymal stromal cells (MSCs) represent a promising therapeutic approach in many diseases, including inflammatory liver disease, in view of their potent immunomodulatory properties. Researchers in MERLIN have looked at the effectiveness of MSCs against inflammatory liver disease in laboratory models, in order to help develop a novel cell therapy for PSC and AIH. A clinical trial was designed based on the outcomes of our studies in the laboratory. The trial is now underway, looking at the safety of the MSC therapy and examining the effect of therapy on inflammation in patients with PSC and AIH. NHSBT have manufactured specially selected MSCs (ORBCEL-C™, discovered by Orbsen Therapeutics) for delivery to patients in the trial. This will set the stage for a potential new, MSC-based therapy in the future. In addition, MERLIN is generating new knowledge that is more widely applicable to MSC therapy in general, informing the development of therapies for other forms of liver disease and other inflammatory conditions.
The main research undertaken and results achieved are described below under the following headings:
• WP1 Pre-clinical efficacy testing of MSC.
• WP2 Functionality of endogenous MSC.
• WP3 Immunogenicity and immunological footprint of MSC in PSC.
• WP4 MoA of MSC in models of liver damage in vivo.
• WP5 Genetic, biological & pharmacological enhancement of human ORB1+MSC re liver inflammation treatment.
• WP6 Advanced tools for bio-distribution of MSC.
• WP7 Generation of Investigational Medicinal Product Dossier and Manufacture of Clinical Grade ORB1+/- MS
• WP8 Phase 2a clinical trial of ORB1+/- MSC in patients with PSC
In addition, the data management and samples management work carried out in WP9, to support the S&T core, are briefly described.
1.3.1 WP1 Pre-clinical efficacy testing of MSC
1.3.1.1 WP1 Introduction
Encouraging results obtained from pre-clinical studies have highlighted the immunomodulatory and clinical potential of mesenchymal stromal cells (MSCs) Consequently, many autologous and allogenic MSC based clinical trials are being undertaken for a number of indications including graft versus host disease (GVHD) and acute and chronic liver disorders. Nevertheless, transplantation of third party MSCs has been shown to be feasible and safe with good tolerability profiles in clinical studies. There are still uncertainties about their long-term therapeutic efficacy. Although some clinical studies have undoubtedly demonstrated clinical efficacy of MSC, most of the studies often yield mixed results, suggesting a significant variability in the immunomodulatory potency of MSC. It is now accepted that the therapeutic potential of MSCs mainly relies on their short-term paracrine ability to reduce inflammation, inhibit immune responses, and produce trophic factors. An important requirement for the clinical usage of MSC would be the production of clinical grade cells with better treatment efficiency. Therefore, in this reporting period in WP1 we set out to investigate whether pre-treatment of ORB+ UC-MSC with either IFN-γ+TGF-β +retinoic acid (RA) or the glucocorticoid steroid budesonide would enhance their immunomodulatory and therapeutic efficacy in the Mdr2Ko/FVB pre-clinical murine model of inflammation mediated liver injury. WP3 and WP5 have previously demonstrated that priming of ORB+ UC-MSC with EMC-cocktail or budesonide enhanced their immunomodulatory potential in vitro. Also, their mechanism of action will be elucidated using mono- and co-culture of MSC with hepatic sinusoidal endothelial cells (HSEC) and biliary epithelial cells (BEC) in vitro.
1.3.1.2 WP1 Key Work Undertaken and Results Achieved
The overall objective of MERLIN is to bring optimised second-generation MSC therapy to the clinic to treat patients with Primary Sclerosing Cholangitis (PSC).
Cellular senescence of hepatocytes, cholangiocytes, stellate cells and immune cells has been implicated in chronic liver disease progression. We have showed that hydrogen oxide can induce senescence (growth arrest) in human biliary epithelial cells. Furthermore, treatment with MSC or MSC conditioned medium reduced biliary epithelial cell senescence adding further to their beneficial effects. In addition, MSC suppressed the proliferation/activation of injurious inflammatory cells (lymphocytes) isolated from both the peripheral blood of patients with PSC and the explanted liver of patients with PSC who have undergone liver transplant. These data confirm the beneficial effects of ORB+ MSC on human effectors of inflammatory liver disease.
Our pre-clinical studies demonstrated that ORB+ MSC from bone marrow and umbilical cord attenuated hepatic inflammation in Mdr2KO/FVB models of biliary injury. Notably, MSC infusion improved liver injury as evidenced by a reduction in serum alanine aminotransferase (ALT) and bile acid (BA) levels, hepatic pro-inflammatory immune infiltrates and the generation of T regulatory cells and anti-inflammatory macrophages. We demonstrated this effect held true for other models of liver injury including Ova-Bil and CCl4.
Our in vitro data demonstrated that MSC could exert their inhibitory effects locally and/or remotely to down regulate leukocyte recruitment. We also described the involvement of IDO in this phenomenon since blocking this enzyme with 1-MT removed the inhibitory effect of not only MSC but also conditioned medium collected from unstimulated/stimulated MSC. This could in turn reduce the expression of VCAM-1 and ICAM-1. Building on this work we were able to demonstrate that MSC could exert similar beneficial effects in vivo after subcutaneous administration suggesting their ability to exert their effects remotely without a requirement to home to the liver.
We sought to establish if we could further improve the effectiveness of MSC by pre-treating them with cytokines. Having tried multiple cytokine combinations we were able to see changes in gene expression but no major differences in terms of their efficacy in vivo in models of liver injury. Therefore, this study suggests that good clinical grade unstimulated ORB+ UC-MSC may substantially contribute to promising clinical outcomes in an appropriate physiological pro-inflammatory micro-environment without any pre-treatments.
1.3.1.3 WP1 Conclusion
This work package has been successful in demonstrating the effectiveness of MSC in reducing liver injury in a number of different models as well as providing evidence of efficacy with human cells. It also provides insights as to the mechanism of action which will be useful in future studies.
1.3.2 WP2 Functionality of endogenous MSC.
1.3.2.1 WP2 Introduction
This work package, WP2, aims to elucidate the functionality of endogenous MSCs during liver injury by significantly depleting their numbers and subsequently monitoring the development of liver injury in specific mouse models. In 2010, Kraman and colleagues developed a conditional knock-out murine model, which allows the depletion of stromal cells, including bone marrow MSCs. These transgenic mice are biologically engineered to express the human diphtheria toxin receptor (DTR) and Luciferase (Luc)2 in cells which naturally express fibroblast activation protein (FAP), a surface protein marker identified on a range of stromal cells, including human and mouse MSCs (Bae et al., 2008; Tran et al., 2013). Administration of diphtheria toxin (DTx) to these mice results in the depletion of any cell expressing FAP via the engagement of the DTR protein. This FAP-DTR murine model should allow us to determine the impact of stromal cell depletion, including populations of MSCs, during liver injury. A limitation of this model is that a range of other non-MSC FAP+ stromal cell populations, identified in muscle, pancreas and lymph node tissues, are also targeted by the DTx treatment, making it difficult to directly attribute any effects observed to the depletion of MSCs. Nevertheless, this initial study could potentially inform future more in depth studies.
1.3.2.2 WP2 Key Work Undertaken and Results Achieved
We have demonstrated that the expression of the FAP protein was significantly increased at the gene and protein level in chronically diseased human liver tissue when compared to normal liver, as confirmed via qPCR and western blotting. Also, we showed the tissue distribution of FAP via immunohistochemical staining is largely restricted to portal areas and fibrotic septa in chronically diseased tissues. Furthermore, we showed that FAP protein expression strongly correlated with the extent of fibrosis in the diseased tissues, as measured by picrosirius red analysis. This was also confirmed at the mRNA level, with FAP mRNA expression exhibiting a strong positive correlation with collagen mRNA expression (Coll IAI). Additionally, in the previous periods, we started to characterise the expression of FAP in murine liver tissues, with initial experiments demonstrating an increase in FAP mRNA expression in the biliary injury model, MDR2-/-, but no significant changes were observed in acute and chronic carbon tetrachloride (CCl4) models of injury.
Finally, we were able to optimise the murine model which utilised diphtheria toxin (DTx)-mediated depletion of FAP+ stromal cells followed by acute CCl4-driven liver injury. However, it became apparent that the depletion of FAP+ stromal cells significantly affected a number of tissues due to the wide expression of FAP in tissues, such as muscle, pancreas and lymph nodes, making it difficult to directly attribute any effects observed to the depletion of MSCs.
Nevertheless, our results suggest that endogenous populations of stromal cells are not simply bystanders during the administration of exogenous MSCs and could potentially play a role in the pathology of both acute and chronic liver injuries. However, it is unclear from the current data whether that role is beneficial or in fact detrimental. Additionally, these data suggest that the effects of depletion are very much dependent on the context of injury type (acute or chronic) and stage (active injury or resolution) and it is evident that further investigation, beyond the scope of MERLIN, is warranted.
1.3.2.3 WP2 Conclusion
We have comprehensively characterised the expression of FAP in murine and human liver injury which will be of significant value to the scientific community. In so doing it has become clear that FAP expression is not limited to just MSC and thus interpretation of knock-down of FAP is challenging. Our data set the scene for further studies that will need to refine this model further if the effect of MSC depletion alone is to be understood.
1.3.3 WP3 Immunogenicity and immunological footprint of MSC in PSC.
1.3.3.1 WP3 Introduction
WP3 developed around 3 key questions concerned with optimisation, mechanisms of action and monitoring of MSC treatment for immunoregulatory therapy. The first question was whether pre-treatment of MSC in vitro could optimise their immunoregulatory efficacy and reduce their immunogenicity. This question was addressed by treating MSC with a variety of factors and determining the effect on MSC immunophenotype, gene expression profile, immunoregulatory function and susceptibility for cytotoxic lysis. We furthermore examined the effect of prolonged culture on MSC phenotype and function and we performed whole genome methylation profiling to examine whether culture-induced changes would leave an epigenetic imprint in MSC.
The second question in WP3 was to elucidate the mechanism of action of MSC therapy. While it was previously thought that the effect of MSC is dependent on the persistence of MSC after infusion, we and others determined that MSC are short lived after infusion. This raises the question of how MSC modulate the immune system after infusion, which we addressed by tracking experiments and measurement of immune parameters after MSC infusion. We also examined whether MSC with optimised immune properties would show a different biodistribution and survival pattern after infusion.
The third question was whether MSC infusion would leave an immunological footprint in PSC patients. To examine this we performed in depth analysis of the immunological footprint in different groups of inflammatory liver disease patients and compared them to healthy controls. We determined immune cell subset distribution and performed serum analyses. These experiments served to setup methodology for measurements of samples from the Merlin clinical trial and provided a background pattern of immune parameters in PSC patients.
1.3.3.2 WP3 Key Work Undertaken and Results Achieved
Experiments were carried out in vitro to treat umbilical cord-derived MSC (ucMSC) with a variety of cytokines, growth factors, vitamins and prostaglandins and examine their immunophenotype, gene expression profile and functionality in T cell suppression assays. Of all tested factors, IFNg was the most potent factor in upregulating the expression of key molecules in MSC involved in their immunoregulatory effect, including IDO and PD-L1, and in enhancing the capacity of MSC to suppress T cell proliferation and IFNg production. At the same time, IFNg containing pre-treatments increased the expression of HLA class I and II molecules on MSC, impacting the susceptibility of MSC for cytotoxic attack. To examine whether pre-treatment would modify the epigenome of MSC, whole genome methylation analysis was performed. Only minor effects on the DNA methylation profile of MSC were detected, indicating that the phenotypical changes induced by pre-treatment were transient. Pre-treated MSC were administered via tail vein injection in a CCl4 inflammatory liver disease mouse model. The differentially treated MSC showed no differences in biodistribution and accumulated in the lungs. There was also no difference in MSC survival and there was no therapeutic effect of control MSC or pre-treated MSC in this model in our hands. Possibly the lack of MSC distribution to the liver prevented a potent effect of MSC on liver inflammation. Therefore we set up an ex vivo liver tissue slice model. In this model MSC treated with a combination of IFNg, TGFb and retinoid acid were most potent in reducing the expression of inflammatory factors after challenging the tissue with LPS (Figure 1).
In addition to treating MSC with growth factors and cytokines, we examined the effect of prolonged culture on the phenotype and function of MSC. Culturing MSC higher passage numbers (from 4-12) reduced their proliferation and their capacity to inhibit T cell proliferation. It was also observed that within a passage MSC showed large differences in DNA methylation patterns from time of seeding until confluency. This was most likely associated with cell density.
Figure 1 Effect of differentially pre-treated ucMSC on ex vivo liver slices. Gene expression levels and cytokine/chemokine levels were measured in liver slices treated with LPS and supernatant, respectively. (a)(b)(c) Both gene expression levels and cytokine secretion levels were measured for MCP-1, TNFα and IL6. In addition, (d) gene expression levels of IP10 and also (e) CXCL1 was measured in the medium were measured. n=6 * p<0.05.
The mechanisms of action of MSC therapy were explored by examining the fate of infused MSC. As explained above, intravenously infused MSC accumulate in the lungs. Double labeling of MSC with Qtracker605 beads, which would disappear from the cells after cell death, and Hoechst, which would be preserved even after cell death, allowed tracking of infused MSC even after death. It was found that directly after infusion the majority of the cells is alive and present in the lungs. After 24h most cells have died and translocate partly to the liver. After 72h no living or dead MSC were detected anymore (Figure 2). The presence of MSC in the lungs was accompanied with elevated numbers of monocytes and neutrophils in lung tissue, which phagocytosed the administered MSC. Monocytes that phagocytosed MSC were subsequently detected in the blood and in liver tissue. Upon phagocytosis of MSC, monocytes adapted a regulatory phenotype and adapted the capacity to induce regulatory T cells. .
Figure 2 Quantitative detection of double-labeled MSC after intravenous infusion in whole mice, and specifically in lungs and liver. The Qtracker signal detects living MSC, whereas the Hoechst signal also detects dead MSC.
To examine the immunological footprint of PSC patients and other inflammatory liver disease patients, blood samples were analysed for immune cell subset frequencies, profiles of plasma cytokines and growth factors and concentration and size of nanoparticles in plasma. We determined naïve, central memory, effector memory and late effector memory CD4 and CD8 T cell frequencies, B cells, NK cells and different types of monocytes and dendritic cells. Distinct immune cell profiles were observed for different groups of inflammatory liver disease patients. PSC patients for instance showed increased percentages of CD19+ B cells compared to other inflammatory liver disease patients and healthy controls (Figure 3). Furthermore, on the basis of plasma cytokine and growth factor profiles and nanoparticle levels distinctions between different patient groups and healthy controls could be made. The deviated immunological footprint of PSC patients and other inflammatory liver disease patients compared to healthy controls indicates that immune profiling of patients in MSC trials may reveal efficacy and safety information.
Figure 3 Frequencies of CD19+ B cells, CD16+CD56+ NK cells and CD14+ monocytes in liver disease patients and healthy donors.
1.3.3.3 WP3 Conclusion
WP3 revealed novel mechanisms of action of MSC therapy. MSC end up in lung tissue after infusion, where they interact with innate immune cells that subsequently adapt a regulatory phenotype and migrate to other sites of the body. Methodology was set up and tested to track MSC after administration and to determine effects of MSC therapy. The immunological footprint of inflammatory liver disease patients was determined, which will serve as a baseline for the Merlin clinical trial.
1.3.4 WP4 MoA of MSC in models of liver damage in vivo.
1.3.4.1 WP4 Introduction
The overall goal of MERLIN is to develop and validate a new treatment option for patients with Primary Sclerosing Cholangitis (PSC) and enhance the functionality of Mesenchymal Stromal Cells (MSC). MSC represent a promising therapeutic approach in many diseases, including inflammatory liver disease, thanks to their potent immunomodulatory properties . MSC inhibit in fact T cell activation induced by an anti-CD3/CD28 antibody stimulus, mitogens, and alloantigens; they also inhibit NK cell activation, as well as B cell terminal differentiation, and dendritic cell maturation and functionality. However the molecular mechanisms responsible for the anti-inflammatory effects of MSC in liver disease are still unknown. Much of the pre-clinical work in liver injury showed that MSC reduce oxidative stress and cellular infiltrate in the liver but it has not been yet defined the mechanism by which this was achieved.
The specific objectives of WP4 were:
1. to identify the environment/locational requirements for MSC-mediated immunomodulation in vivo
2. to identify the molecule/s responsible for the anti-inflammatory effects of MSC in vivo
3. to identify the action of MSC on the hepatic vasculature in suppressing inflammation
1.3.4.2 WP4 Key Work Undertaken and Results Achieved
In order to identify the environment/locational requirements for MSC-mediated immunomodulation in vivo we exploited MSC encapsulation that provides conclusive data in support of either the endocrine or the paracrine/contact signalling required for MSC immunomodulation . Thus, we injected alginate-encapsulated human CD362+ MSCs (E-MSCs) in a mouse model of local inflammation induced by s.c. injection of CFA/OVA . Our results clearly showed that BM CD362+MSCs exert their immunomodulatory properties in vivo, even better than unsorted MSCs, but encapsulation may be detrimental in the case of human, sorted MSCs. (Figure 4)
To formally demonstrate that s.c. injected WT or CD362+MSCs do not leave the injection site, in collaboration with Bioinvision, we performed whole-mouse cryo-imaging analysis of mice immunized with CFA/OVA and injected with fluorescently labelled hMSCs, confirming that neither WT nor CD362+MSCs migrated away from the site of injection and are able to dampen inflammation through the release of soluble mediators.
In an effort to understand the molecular mechanisms responsible for the endocrine effects of MSC, we exploited a standardised workflows previously developed in our laboratories for shotgun proteomic characterization of the mouse MSC secretome . Our results on MSC secretome indicated that, upon activation by inflammatory cytokines, MSC upregulate the expression of several proteins potentially affecting angiogenesis and inflammation through multiple pathways. Interestingly, when we compared the secretomes of human and mouse MSCs, we found that only 16 proteins are upregulated in both cell types and 11 of them modulate angiogenesis directly or indirectly, thus supporting the idea that the endothelium is a specific target of MSC during inflammation. Among the molecules upregulated upon cytokine stimulation, we focused our attention on the tissue inhibitors of metalloproteinases 1 (TIMP1). Thus, we assessed the role of TIMP-1 secreted by MSC in the CFA/OVA mouse model of local inflammation, demonstrating that MSC inhibit high endothelial cell activation and lymphocyte homing to lymph nodes by releasing TIMP-1 and further identifying the endothelium as a specific and novel target of MSC-based therapy. (Figure 5). Therefore, we proposed that pre-activation with pro-inflammatory cytokines strengthened the anti-angiogenic effects of MSC, thus supporting our hypothesis that, during an inflammatory response, MSC target angiogenesis and thus dampen the inflammatory response. Additionally, in agreement with data obtained with mouse MSCs, soluble factors secreted by human MSCs affected the tubulogenesis of HUVEC cells . However, in the case of human cells, MSC-conditioned media (CM) was able to inhibit tube formation even when MSC had not been primed by cytokines. Of note, in the case of human MSC, TIMP-1 blockade restored the formation of the HUVEC endothelial network in the presence of either unstimulated (unstim) or stimulated (stim) human MSC-CM, thus pinpointing TIMP-1 as a crucial MSC-derived modulator of angiogenesis also in human models. (Figure 6)
To investigate the role of MSC in modulating angiogenesis in inflamed livers, we performed tube formation assays of human sinusoidal endothelial cells (HSECs) from PSC patients in the presence of CM from unstimulated or cytokine-stimulated bone marrow human MSCs. Of note, both unstimulated and stimulated MSC-CM induced a significant reduction in tube formation of HSECs in comparison with control medium (aMEM). Further, we assessed the contribution of TIMP-1 in inhibiting angiogenesis of HSECs in the presence of human MSC-CM. To this purpose we performed tube formation assays of HSECs by adding anti-TIMP-1 blocking antibody to CM from both stimulated and unstimulated bone marrow hMSCs, using the appropriate isotype as control (Figure 7). Our results clearly show that TIMP-1 plays a fundamental role in the inhibition of angiogenesis also of liver sinusoidal endothelial cells.
Further, we investigated the contribution of TIMP-1 in controlling liver inflammation, by overexpressing TIMP-1 in MDR2 -/- mice by AAV9-mediated gene transfer and analyzed liver immune cell infiltration. AAV-TIMP-1 injection resulted in reduced liver infiltration of myeloid cells and did not affect Kupffer cells population. Moreover, TIMP-1 overexpression reduced the amount of neutrophils and infiltrating macrophages and increased the amount of restorative macrophages in comparison with control untreated MDR2 -/- mice. Further, AAV-TIMP-1 treated mice showed lower levels of bile acids, ALT and ALP than untreated MDR2 -/- mice (Figure 8), suggesting a beneficial effect of TIMP-1 overexpression in PSC mouse model. However, the AAV-LacZ injected mice used as AAV-infection control collectively showed less injury than untreated MDR2 -/- mice, suggesting that the observed effects were not specifically dependent on AAV9-TIMP-1. We also found that high levels of TIMP-1 negatively correlate with Treg accumulation in injured livers of MDR2-/- mice. Thus, we blocked TIMP-1 by administering an anti-TIMP-1 blocking antibody in MDR2 -/- mice. Nonetheless, TIMP-1 blockade did not result in amelioration of clinical score of liver injury, thus indicating that, although TIMP-1 plays an important role in dampening liver immunosuppression, other factors might contribute to liver injury in MDR2-/- mouse model (Figure 9).
Finally, we assessed the effects of extracellular vesicles (EVs) derived from human BM MSC in our PSC mouse model, finding that EVs from MSC might improve the clinical outcome of MDR2-/-mice.
Figure 4 Effect of hMSC on OVA-specific T cell proliferation. The experimental protocol was designed to investigate the influence of hMSC upon the activation of OT62 (OVA6specific) T cells: OT2 CD45.2 CFSE6labelled cells (1x106) were transferred into CD45.1 C57 mice. After 24 hrs mice were immunized with OVA peptide (100 ng) in complete Freund s adjuvant (CFA) by s.c. injection into the footpad. On day 2 a group of animals received 500.000 hMSC wt injected s.c. in the back another group received 500.000 hMSC S+ s.c. and the last group received 500.000 encapsulated (E6)hMSC S+ s.c.. Immunized mice were sacrificed at day 4 the popliteal draining LNs (dLNs) were collected and cells were analysed by flow cytometry.
Figure 5 TIMP-1 mediates the anti-angiogenic effect of MSC-CM in vitro and the anti6inflammatory effect of MSC in vivo.SVEC4-10 network formation in matrigel in the presence of MSC6CM or unst MSC6CM and anti mTIMP1 blocking antibody. A) Representative images at 6 hrs (left). Segment length quantification (right). Data are expressed as mean ± SEM (*p<0.05 **p < 0.01; One way Anova). B) Diagram of the experimental protocol designed to investigate the contribution of TIMP61 to the anti6inflammatory effects of MSC. Mice were immunized in the dorsal back with CFA/OVA on day 0 and, on day 1, 3 groups of animals received subcutaneous injection of 106 MSC in the lumbar back. 18 hours after MSC transplantation goat polyclonal anti-TIMP-1 IgG or isotype-matched goat IgG was i.v. administrated. On day 4 brachial LNs were collected, processed and analysed by flow cytometry; C) Graph showing the absolute number ofCD456CD31+ cells per single LN expressed as normalized percentage on CFA/OVA (3independent experiments, n=28 dLN). D) Diagram of the experimental protocol designed to over6express TIMP61 in immunized mice. One day after AAV9 administration (day 0),mice were immunized with CFA/OVA. Brachial draining LNs (dLNs) were collected 4days after immunization and digested for FACS analysis. E6G) Graph showing LNcellularity, absolute count of CD45+ and CD456CD31+ cells per single LN expressed as normalised percentage on CFA/OVA. In E6G error bars represent standard error, n= 6 mice for each condition (*p < 0.05; **p < 0.01; t-test
Figure 6: Timp-1 blocking reverts the anti-angiogenic effect of mouse and human MSC conditioned media. MSC-derived TIMP-1 concentration in (A) mouse and (B) human unstimulated or stimulated MSC conditioned medium was measured with ELISA. Data are expressed as mean ± SEM (*p<0.05 parametric T Test). Tube formation assay was performed in the presence of (C) mouse or (D) human TIMP-1 blocking antibody. Representative images of (C) SVEC4-10 cell line or (B) Huvec cells are taken with a phase contrast inverted microscope at 4× objective magnifications. Graphs show the quantification of the tube segment length measured with ImageJ Angiogenesis Analyzer. Data are expressed as mean ± SEM (*p<0.05 **p<0.01; One way Anova).
Figure 3
Figure 7
Figure 8
Figure 9 TIMP-1 modulates regulatory T cells accumulation by regulating TGF beta activation in MDR2-/- mouse livers. The role of TIMP-1 in modulating liver inflammation was assessed by using anti-TIMP-1 blocking antibody. Experimental schedule is represented in the panel (A). 6-week-old MDR2-/- mice were i.p injected twice a week with anti-TIMP-1 blocking antibody, using isotype as control. Mice were sacrificed 7 days after the first injection and liver and blood were harvested. Livers were mechanically and enzymatically (Collagenase II and DNAse I 0,2mg/ml at 37°C for 1hour) digested for FACS analysis of immune cell infiltration. Graphs show regulatory T cells (CD45+ CD3+ CD4+ CD25+ FoxP3 + within the CD45+ leukocytes) (B) and amount of active TGF beta, measured by ELISA (C). In the panel D, IHC for CD31, marker of endothelial cells, show the altered liver tissue in MDR2-/- mice. Blood was processed to obtain serum and analysed for the quantification of Bile Acid (E), ALT (F), and ALP (G). n= 4 mice per group, data are expressed as mean ± SEM (*p<0.05 **p<0.01 One way Anova).
1.3.4.3 WP4 Conclusion
Exploiting a mouse model of local inflammation, we demonstrated that MSC inhibit high endothelial cell activation and lymphocyte homing to lymph nodes. In order to clarify MSC mechanism of action, we performed a high-throughput analysis of MSC secretomes that provided us a plethora of pharmacologically targetable molecules for the treatment of inflammation. Among the upregulated proteins, we focused our attention of TIMP-1, providing evidence that TIMP-1 secreted by MSC plays a fundamental role in the inhibition of angiogenesis in vitro and in vivo, inhibiting T-cell homing into inflamed lymph nodes.
Importantly, we found a common antiangiogenic signature in the secretome from both mouse and human MSC stimulated with proinflammatory cytokines. In agreement with data obtained with mouse MSCs, we also showed that soluble factors secreted by human MSCs affect the ability of HUVEC cells to form tubes. Interestingly, in the case of human cells, MSC-CM was able to inhibit tube formation even when MSC had not been primed by cytokines. However, pre-activation with pro-inflammatory cytokines strengthened the anti-angiogenic effects of MSC-CM, thus supporting our hypothesis that, during an inflammatory response, licensed MSC target angiogenesis and thus dampen the inflammatory response.
Collectively, our results clearly show that TIMP-1 plays a fundamental role in the MSC-mediated inhibition of angiogenesis in vitro and in vivo model of local inflammation. We confirmed our findings by exploiting mouse and human MSC, as well as both mouse ad human endothelial cells. Importantly, we found that TIMP-1 plays a fundamental role in the inhibition of angiogenesis also of liver sinusoidal endothelial cells from PSC patients. However, TIMP-1 AAV9 mediated gene transfer in MDR2-/- mice resulted in only a partial beneficial effect. Of note, this effect seemed also to be independent from TIMP-1 and likely associated with AAV vectors injections.
Nevertheless, although high levels of TIMP-1 negatively correlate with Treg accumulation in injured livers of MDR2-/- mice, TIMP-1 blockade did not result in amelioration of clinical score of liver injury in MDR2-/- mice, thus indicating that other factors might contribute to liver injury in this mouse model. Finally, we found that extracellular vesicles (EVs) derived from human BM MSC might improve the clinical outcome of MDR2-/-mice.
1.3.5 WP5 Genetic, biological & pharmacological enhancement of human ORB1+MSC re liver inflammation treatment.
1.3.5.1 WP5 Introduction
The overall objectives of WP5 were:
• To isolate, expand, and cryopreserve CD362+MSC and PA-MSC from human bone marrow and umbilical cord tissues under GLP conditions.
• To provide GLP-compliant MERLIN SOPs and BMRs to partners to define, govern and record the thawing and expansion of cryopreserved human ORB1+MSC and PA-MSC in each WP laboratory.
• To assess the impact of thawing/washing of cryopreserved stromal cells upon efficacy in CCl4-induced model of liver Injury.
• To assess the impact of prolonged cell culture passage upon MSC efficacy in vitro and in CCl4-induced model of liver Injury.
• To determine the impact of ORB1-overexpression upon MSC efficacy in CCl4-induced model of liver injury.
• To explore unmodified and ‘optimised’ ORB1+ MSC in CCl4-induced model of liver injury.
1.3.5.2 WP5 Key Work Undertaken and Results Achieved
In Period 1, WP5 focussed on the manufacture of cryopreserved CD362+ MSC and PA-MSC from human bone marrow and umbilical cord tissue - and delivered cells to partners in WPs 1, 3, 4, 5 and 6 for testing. >200 vials of cryopreserved CD362+, and PA-MSC from human marrow and umbilical cord were delivered to MERLIN partners for pre-clinical experiments.
In addition, studies were completed that show no impact of cryopreservation/thawing on CD362+ MSC efficacy in the CCl4-induced model of liver injury in the C57Bl6 mouse.
The impact of the glucocorticoid Budesonide on the efficacy of ORB1+ MSC in suppressing T lymphocyte proliferation was also assessed. Pre-treatment of CD362+ MSC with Budesonide stimulates expression of the CD362 protein, GILZ mRNA and TIMP1 RNA. Notably, Budesonide-treated ORB1+ MSC significantly improved MSC-mediated suppression of T lymphocyte proliferation.
In total ORB delivered over 400 vials of cryopreserved stromal cells to MERLIN partners over the course of Period 1 and Period 2. ORB explored novel activators of ORBCEL-C and focused on developing Budesonide as an activator of ORBCEL-C.
Researchers have worked on enhancing the manufacturing of the ORBCEL-C product using the MACSQuant Tyto to select near 100% pure stromal cells and expanding ORBCEL-C on the closed automated Quantum Bioreactor and this led to a collaboration between ORB and NHBST to test Quantum expanded ORBCEL-C in the REALIST clinical trial (https://clinicaltrials.gov/ct2/show/NCT03042143)
ORB have explored the potency of ORBCEL-C with the CD4 T cell proliferation assay and VCAM1-specific Endothelial cell activation assay and discovered a biomarker (ORB2) that allows ORB to select effective cord tissue donors for ORBCEL-C Manufacture.
ORB collaborated extensively with NHSBT and UOB over the entire MERLIN project to transfer the ORBCEL-C Manufacturing process to NHSBT for the MERLIN Phase 2a clinical trial. This has allowed the MERLIN trial to progress to treat its PSC patient in March 2019.
https://www.prnewswire.co.uk/news-releases/clinical-trial-investigates-orbsen-therapeutics-orbcel-c-tm-immunotherapy-as-treatment-for-two-autoimmune-liver-diseases-820613487.html
Finally, ORB have shown that ORBCEL-C derived extracellular vesicles express CD362, CD39 and CD73 and retain intrinsic ATPase activity. With a view to understanding the consistency of the ORBCEL-C product, ORBCEL-C derived from 7 donors in 8 series of experiments was tested in a rat model of lung inflammation and showed consistent function of the ORBCEL-C Drug Product.
1.3.5.3 WP5 Conclusion
MERLIN WP5 enabled ORB to develop the ORBCEL-C selection and GMP manufacturing process from the pre-existing ORBCEL-M marrow CD362-selection & expansion process developed in the REDDSTAR program. ORB then manufactured and distributed >400 vials of ORBCEL-C for pre-clinical testing, optimisation and in vitro evaluation in WP1, WP3, WP4, WP5 and WP6. ORB have tech transferred the ORBCEL-C Manufacturing process to NHSBT in WP7 to allow clinical manufacture of ORBCEL-C for the MERLIN clinical trial in WP8. ORB have established the 2 year shelf life of the ORBCEL-C Drug Product. ORB have since established GMP manufacturing operations and are collaborating with UOB to design a Phase 2b MERLIN2 Adaptive Licensing in patients with autoimmune liver disease.
1.3.6 WP6 Advanced tools for bio-distribution of MSC.
1.3.6.1 WP6 Introduction
BioInVision, based in Cleveland, Ohio, USA, offers imaging instrumentation and methodologies critical to preclinical studies. The unique CryoVizTM instrument, utilizing the patented cryo-imaging technology, allows microscopical anatomical and molecular fluorescence imaging of laboratory small animals such as a mouse or organs excised from them with single-cell sensitivity. With its sub-ten-micron-scale imaging, cryo-imaging allows one to detect even single stem or cancer cells anywhere in a mouse. The technology is also offered as a service and is targeted to a variety of biomedical applications including stem cell homing and biodistribution, cancer metastasis, imaging agents, drug discovery and delivery, tissue engineering, mouse phenotyping etc. BioInVision’s cryo-imaging technology was a good match for the MERLIN project which focused on pre-clinical and clinical studies to investigate specifically the use of MSCs as a therapy for Primary Sclerosing Cholangitis (PSC), a type of liver disease for which there is currently no cure. In MERLIN, BioInVision partnered with several EU institutions developing stem cell-based therapies specifically targeting the inflammatory components of liver disease. BioInVision’s cryo-imaging technology powered by the CryoVizTM instrumentation and software was employed to provide high-resolution, microscopic colour anatomical and molecular fluorescence volumetric imaging, accurate stem cell detection, manual review and editing of cell detection, probabilistic segmentation of organs in 3D from anatomical and fluorescence volumes, global and local cell quantification, and 3D visualization of stem cell biodistribution within mouse-sized volumes. Together, MERLIN consortium partners and BioInVision set out to develop MSC therapy for liver disease.
1.3.6.2 WP6 Key Work Undertaken and Results Achieved
For MERLIN, BioInVision customized the CryoVizTM instrument to support (i) high throughput cryo-imaging of multiple-sample-arrays within the same frozen block, (ii) smart image acquisition that employs a tissue detection algorithm for skipping over unnecessary regions during imaging thereby reducing imaging time, (iii) multi-fluorophore stem cell detection that enabled tracking of multiple populations of stem cells as well as immune cells within the same specimen, and (iv) semi-automatic organ/tissue segmentation that aided in analysing local biodistribution of cells within specific organs and tissues. In software, BioInVision developed the MERLIN Biodistribution Explorer (MBE), a tool for analysing large volumes of cryo-imaging data, thereby complementing the hardware modifications to the CryoVizTM. MBE supports profile-based access privileges upon login, structured browsing of experimental data organized by work package, institution, lead investigator, and imaging study, browsing of metadata, 3D graphics and movies, and text comments relevant to an imaging study, and live data viewing in 2D/3D in an interactive fashion. MBE is augmented by an excellent set of software modules that provide high-throughput stem cell detection, manual review and editing of detection results, global and local quantification of cells, probabilistic 3D segmentation of organs and tissue, and 3D visualization of stem cell biodistribution within mouse-sized volumes. Together, the modified CryoVizTM system and software were employed in several work-packages of MERLIN, which we describe in the next paragraphs.
First, in a study with UNIPD, CryoVizTM was used to investigate the mechanism of action of MSC (WP4). Mesenchymal stem cells (MSC) represent a promising therapeutic approach in many diseases in view of their potent immunomodulatory properties, which are only partially understood. Although MSC seem to be effective in treating diverse immune-mediated disorders, a unifying mechanism of action was missing. In a previous study, the proteomic analysis of the MSC secretome identified the tissue inhibitor of metalloproteinase-1 (TIMP-1) as a potential effector molecule responsible for the anti-angiogenic properties of MSC. Using CryoVizTM cryo-imaging and cell dispersion analysis module within MBE, it was verified that subcutaneously injected MSC did not migrate away from the site of injection during the experimental time (5 days), both in immunized and in untreated mice (Figure 10). Together with the previous study, these data indicate that MSC are able to dampen inflammation through the release of soluble mediators. This study of the broad anti-inflammatory properties of MSC paves the way for developing strategies that exploit MSC-mediated inhibition of lymph node angiogenesis in the treatment of inflammation-associated pathologies.
Second, in collaboration with EMC, the immunogenicity of MSC was examined (WP3). Mesenchymal stromal cells (MSC) possess immunomodulatory properties and are low immunogenic, both crucial properties for their development into an effective cellular immunotherapy. They have shown benefit in clinical trials targeting liver diseases, however the efficacy of MSC therapy would benefit from improvement of the immunomodulatory and immunogenic properties of MSC. In this cryo-imaging study, the responsiveness of ucMSC to in vitro optimisation treatment was studied. By employing 3D visualization and quantification analysis of biodistribution from whole mouse cryo-imaging volumes, we observed that four hours after intravenous infusion in mice with CCl4-induced inflammatory liver injury, the majority of ucMSC were still trapped in the lungs. Rapid clearance of ucMSC(VitB6), ucMSC(Starv+VitB6) and ucMSC(MC) and altered bio-distribution of ucMSC(TGFβ) compared to untreated ucMSC was also observed (Figure 11). This study confirmed that majority of infused ucMSC accumulate in the lungs and that there is a rapid clearance of cells, with the cells mostly moving to the liver, an important finding that would inform future cellular immunotherapy strategies.
Third, in partnership with UoB, the efficacy and biodistribution of MSC in both acute and chronic mouse models of PSC was studied (WP1). The ability of mesenchymal stromal cells (MSC) to immunomodulate offers therapeutic potential but the inherent heterogeneity of unsorted MSC populations may explain varied/reduced function and poses regulatory challenges. Thus, in this study, the efficacy of purified CD362+ MSC in relevant pre-clinical models of PSC as a prelude to a clinical trial was examined. A time-course of bio-distribution was determined by whole mouse cryo-imaging of Q-dot labelled MSC, followed by MBE analysis. A seven-day time-course indicated MSC were cleared rapidly although there was a liver-specific increase in MSC 2-3 days after infusion (Figure 12). We concluded that CD362+ human MSC exert a potent anti-inflammatory action in murine models of PSC.
Last, in partnership with ORB, homing and biodistribution of ORB1+ MSC was studied upon injection into healthy and CCl4-induced liver injury mice using CryoVizTM cryo-imaging (WP5). First, ORB1+ MSC were isolated, expanded, and cryopreserved. Subsequently, they were injected into the two groups of mice - healthy and injured - and their homing and biodistribution was studied using MBE software. In the first experiment, mouse MSC from UBC-eGFP mice (green) were injected into FAP-DTR BAC Tg mice (low and high doses) to study homing and distribution. The aim of the experiment was to see if ablating FAP+ cells with DTX in the FAP-DTR mouse altered the engraftment distribution of the GFP+ MSC. In a second experiment, PKH26 (red)-labelled MSC were injected in uninjured and acute CCL4 injury mice to study distribution, homing and survival. It was found that cell retention and homing to affected organs was higher in the injured group as compared to the control group. (Figure 13).
Figure 10
Figure 11
Figure 12
Figure 13
1.3.6.3 WP6 Conclusion
The MERLIN project aimed to develop a therapy using purified MSC populations that alleviates the ongoing inflammation that drives fibrosis in liver. Prior to MERLIN, there was lack of evidence and concerns regarding human PA-MSC survival and persistence upon transplantation into mice. The research faced unknowns such as retention period of MSC anti-inflammatory behaviour, whether there was differentiation upon infusion, the molecular mechanisms underpinning efficacy, immuno-modulatory action in vivo, etc. Pre-clinical imaging experiments on the CryoVizTM conducted by BioInVision along with MERLIN partners through several work-packages (WP1- UoB, WP3- EMC, WP4- UNIPD, WP5-ORB) have helped provide some answers, especially pertaining to MSC homing and biodistribution upon transplantation into injured and healthy mice. This knowledge is enabling the MERLIN consortium partners to better understand the mechanism of action of the MSCs in mouse models. The study results will help optimize the MSC type to be administered, develop cell preparation methods, and optimize the dosage strategy. All the pre-clinical findings will enable a successful Phase 1/2 clinical safety trial in patients with PSC.
1.3.7 WP7 Generation of Investigational Medicinal Product Dossier and Manufacture of Clinical Grade ORB1+/- MS
1.3.7.1 WP7 Introduction
The aim for NHSBT has been to develop a GMP compliant process for the manufacture of clinical grade CD362+ve enriched Mesenchymal Stromal Cells from umbilical cord tissue and provide these cells in approved batches for a clinical trial. This work has included completion of development and engineering runs plus validation of standardized QC testing and final release criteria in place. NHSBT has also undertaken a stability study on the Drug Product. The NHSBT manufacturing facility in Birmingham, UK, has an MHRA license for this work and the results informed the Clinical Trial Application which included the development of an IMPD and Investigator Brochure. The Drug Product is produced directly from the Drug substance and cryopreserved in aliquots of 40 x10^6 or 80 x10^6 cells per bag, ready for issue and transport to the ward and then thawing ahead of administration to patients.
1.3.7.2 WP7 Key Work Undertaken and Results Achieved
From the start of the grant NHSBT worked closely with the other partners on the pre-clinical development and small-scale cell culture work as part of a technology transfer process to enable future production of MSCs under GMP by NHSBT. The NHSBT manufacturing unit along with the NHSBT Quality Assurance Dept reviewed the process design, equipment, proposed consumables and reagents to ensure that a full-scale process could be developed in compliance with MHRA regulations. Optimisation of QC assays, preparation of SOP and BMR documentation and equipment Installation and Operational Qualification validation studies were carried out.
Extensive work was undertaken to develop a consistent supply of Bone Marrow as a source of MSCs. As this proved challenging, the decision was taken to switch the source material for MSCs to umbilical cord (UC) tissue of which NHSBT has ready access to via the NHS Cord Blood Bank.
The development of a cell selection protocol to enrich for CD362+ cells from UC was undertaken in the first period although this subsequently required additional work to ensure the process met GMP standards. A significant achievement in period 2 was the development of a complete process for MSC production from UC tissue that could achieve MHRA approval. The development and validation of this process formed the bulk of the contribution from NHSBT towards the preparation of the initial IMPD and the Clinical Trial Application.
The manufacturing process required extensive development for the cell selection and cell expansion steps. This resulted in some delays and in particular, the process of selecting viable CD362+ cells from umbilical cord tissue proved difficult to develop under GMP conditions, with many of the materials not being available as CE marked for clinical use. In addition, a problem was later identified with a cell dissociation device used to assist the enzyme digestion of the UC tissue. Therefore, further development and validation work was undertaken during P3 to ensure that a new rocker device could be used in a temperature controlled incubator as a suitable alternative for tissue digestion ahead of the subsequent CD362+ cell enrichment. Thus, while important challenges were encountered, they were ultimately overcome.
Methodologies for testing of the Drug Substance and Drug Product ahead of QP release were finalised during P3. The testing included Endotoxin testing, cell counts and morphology for the Drug Substance and flow cytometry for MSC phenotype, sterility, mycoplasma testing and karyotyping for the Drug Product.
As the overall process to produce cells for the clinical trial was delayed a no-cost extension was agreed to run for a year from January 2018.
Following the problems with tissue dissociation a number of additional full-scale validation runs were undertaken using the new method in the GMP compliant facilities of NHSBT Birmingham. A stability programme was also developed for the Drug Product and for the transport of the Drug Product to the clinical units.
The problems identified with the tissue dissociation device and the need to develop another process for this stage of manufacturing meant there was a need to submit a revised IMPD to the MHRA. Once the approval for the amendment was received from the MHRA the manufacture of clinical grade MSCs for use in the clinical trial was finalised and NHSBT then started to prepare cryopreserved Drug Product batches with in-house QP release for clinical use from June 2018. The process takes many weeks to expand sufficient cells from the low numbers of CD362+ enriched cells derived from umbilical cord tissue. Each batch is produced from final cell cultures in multiple Cell Stack flasks after Passages through a number of smaller cell culture plates and flasks. The Clinical trial opened in late 2018 for patients with PSC and AIH and will therefore run beyond the end of the P4 extension period. NHSBT has produced many of the batches needed for the trial during P4 and is committed to producing the final batches required in 2019.
1.3.7.3 WP7 Conclusion
The Clinical trial opened in late 2018 for patients with PSC and AIH and so the primary endpoints of the grant have been met. A number of difficulties were encountered during the grant and lessons learned by NHSBT are primarily focussed on the difficulties in translating an R&D process into GMP compliant methodology. Many of the R&D starting procedures did not lend themselves easily to a standard GMP process that can be undertaken by a range of staff within a GMP facility. In addition, many of the essential reagents did not have CE marking and confirming these could be used under GMP was a time-consuming process. Linked to this, the later failure of a key piece of equipment to function effectively under GMP delayed the project extensively and resulted in the need to repeat many of the process development steps and validations. The final process developed is however now delivering the Drug Substance and Drug Product batches `required although it is not straightforward and will not lend itself easily to any future scale up.
The MERLIN project has been instrumental in creating an experienced team of staff within NHSBT capable of undertaking similar early phase tech transfer and development projects more readily in the future. The team has become more experienced and knowledgeable both in terms of GMP compliant manufacturing and also in the requirements of the Clinical team to gain success in an early phase clinical trial of a novel cell therapy.
1.3.8 WP8 Phase 2a clinical trial of ORB1+/- MSC in patients with PSC
1.3.8.1 WP8 Introduction
The ultimate goal of the MERLIN project is an open label clinical trial delivering CD362 selected umbilical cord derived, mesenchymal stromal cells, to patients with immune mediated liver disease. The clinical trial team remains ready to start this process, which is always challenging. Fundamentally this WP has remained live and with all the components ready to start the trial but technical challenges have delayed our progress. These technical challenges are inevitable when one proposes a novel experimental medicine study of a cell therapy, such as is planned. Participant safety is paramount and therefore the trial must be on hold until the manufacture of clinical-grade product is complete and all related regulatory approvals are in place. This appears to be close which is inevitably exciting for all.
1.3.8.2 WP8 Key Work Undertaken and Results Achieved
We generated the relevant regulatory paperwork for the Merlin trial including the study protocol, culminating in formal MHRA and Ethics submissions. D8.1 (Clinical Trials Authorisation document for the Medicines and Healthcare Products Regulatory Agency (MHRA)) was submitted in Period 2. This included an updated approval from the MHRA, in view of changes to the protocol.
The clinical trial is ongoing; the first patient was treated on 29 March 2019.
Figure 14 Press Release for the Start of Clinical Trial (UoB)
1.3.8.3 WP8 Conclusion
WE were able to overcome the many challenges of obtaining regulatory approval for the trial and successfully manufacturing MSC and opened the clinical trial in Dec 2018. We have recruited multiple patients and treatment thus far has gone well with no adverse effects. We will support the delivery of the trial with other resources and look forward to completing the trial by late 2020.
1.3.9 WP9 Data and Samples Management
1.3.9.1 WP9 Introduction
The multi-disciplinary nature of MERLIN meant that many different types of data were created and analysed throughout the project. To ensure that data were securely but openly shared, and that the movement of samples was efficiently tracked between partners, we established suitable online infrastructure, based on PT’s StudyVault platform technology.
1.3.9.2 WP9 Key Work Undertaken and Results Achieved
Data Management
During Period 1, working with the MERLIN data partners, a bespoke customisation of StudyVault was established for use in the project - StudyVault-MR. PT technologists worked with the data partners on the co-design (task 9.1) configuration (task 9.2) and installation and validation (task 9.3) of the system. On-going software support was also provided (task 9.4). The specific and unique requirements of MERLIN meant that very substantial customisation of the system was required, with some adjustment of the underlying architecture and data handling models.
StudyVault was established online and accessed via https://www.studyvault.eu/. All research data was held in an encrypted format on the cloud (there was no personal patient data created or shared on the system). Data security and confidentiality were key features. While all MERLIN data was pre-clinical research data, and so no personal or human-related data was stored, StudyVault is engineered to support compliant, secure storage of clinical data.
Co-design, installation and validation of the bespoke system was completed in the first years of the project. On-going support and any adjustments were delivered using a continuous delivery approach (so adjustments and customisations of the data platform did not hinder the performance of the live system). Significantly we added support for large-volume data formats from molecular biology platforms.
Key deliverables D9.1 Data Platform Configuration Report and D9.2 Data Platform Verification Report were completed and submitted in period 1.
During periods 2 to 4, the WP9 team continued to provide support to data partners, uploading additional data as required and making improvements to the StudyVault-MR system.
Figure 15 Example StudyVault screen
Samples Management
The management and monitoring of the movement of samples from one partner facility to another is a critical process, which has been found to frequently cause issues in life sciences projects. A key aim of WP9 was to marshal the samples management process and ensure smooth delivery of samples requirements.
Following liaison between the partners most involved in the creation and shipping of samples, an infrastructure for samples management in MERLIN was established in period 1.
During period 2, samples were transferred and managed for pre-clinical studies and sample transfers proceeded smoothly. There was a review of samples transfer logistics and some changes were introduced. The MERLIN Master Samples Register was produced. Partners opted for direct contact as the central approach for samples management, with an additional layer of oversight provided by collecting samples transfer data (with scope for partners to highlight any problems). A samples subgroup was established within the consortium to deal with any issues arising.
Samples during period 3 were dealt with and managed for pre-clinical studies without any issues arising. Direct contact between partners remained the central approach. The Master Samples Register for period 3 captured details of samples transfers, as well as protocols for sample preparation prior to transfer, storage on arrival and handling prior to use.
The Master Samples Register was maintained throughout RP4. Throughout the project, the management of samples and their transport went smoothly, and common issues of samples tracking were avoided.
1.3.9.3 WP9 Conclusion
The establishment and maintenance of the StudyVault system addressed the key challenge of multi-format, multi-owner data management in MERLIN. The system provided the facilities and supports required by the life-sciences partners, while maintaining data security and integrity.
The samples management system also worked as planned, and ensured that MERLIN avoided common research project issues and delays around ownership, location and status of samples.
1.3.10 Main S&T Results: Conclusion
As will be evident from the above, MERLIN has generated new knowledge about the efficacy, mechanisms of action, immune response, optimisation and bio-distribution of MSCs. The project has delivered important data and lays the groundwork for the future tailored production and use of MSCs with precise characteristics for particular applications.
We have also developed advanced technologies for 3D imaging and a novel GMP-compliant manufacturing process for producing MSC-based therapeutic product (ORBCEL-C™).
We are currently in the process of delivering an early phase clinical trial, exploring the safety and effect of our therapy on inflammation in patients with PSC and AIH. The MERLIN trial will continue after the formal end of the project.
We believe the project will ultimately impact on future research, manufacturing of MSCs for clinical use and cell therapies for patients with PSC and other inflammatory diseases.
Potential Impact:
In MERLIN, WP10 was dedicated to maximising the value of the project by (i) communicating the project concept, progress and findings to a range of audiences and (ii) focusing on the future exploitation and commercialisation of MERLIN project results
1.4.1 Impact and exploitation
1.4.1.1 Overview
The MERLIN project has been looking at new ways to treat liver disease with mesenchymal stromal cells (MSCs) sourced from umbilical cord, specifically focusing on PSC and AIH as model diseases. As part of our work the team has looked at the effectiveness of MSC against inflammatory liver disease in laboratory models, explored mechanisms of action of MSC and studied optimum conditions for MSC production. We have also developed advanced technologies (3D imaging and processes for GMP-compliant manufacturing of MSCs).
Our work in the MERLIN project has culminated in the design and launch of a clinical trial, looking at safety and at the effect of MSC therapy on inflammation in patients with PSC and AIH. We hope that the MERLIN clinical trial will set the stage for future MSC-based therapies for PSC, AIH and other conditions involving inflammation.
1.4.1.2 Key Results
Some of the most significant findings from our results in the MERLIN project and our key achievements are set out below.
Efficacy:
• MSC reduce markers of liver damage and inflammation in pre-clinical models of inflammatory liver disease, as well as the number of inflammatory cells in damaged areas. MSC administration ameliorates PSC pathophysiology. Both bone marrow (BM) and umbilical cord (UC) derived MSCs show positive effects.
• MSC suppress inflammatory cells (lymphocytes) isolated from the peripheral blood and explanted livers of patients with PSC.
• Freshly thawed MSC are as effective as MSC used fresh from culture, or freshly thawed and washed (in murine models of liver damage).
• The route of infusion of UC MSC has no impact on therapeutic efficacy (MSC infused subcutaneously and intravenously were equally effective).
Mechanism of Action:
• MSCs injected under the skin do not migrate to other locations. Thus, the effects of MSC may be due to cytokines released by the cells (soluble mediators).
• The beneficial effect of MSCs on PSC is due, at least in part, to the extracellular vesicles (EVs) that MSCs secrete. Results suggest MSCs act as sensor of the surrounding inflamed environment, secreting EVs that block angiogenesis, limiting inflammation. The glycoprotein TIMP-1 plays a key role in the inhibition of angiogenesis by EVs.
• MSCs are phagocytosed (engulfed) by monocytic cells that subsequently migrate to other sites via the blood stream. Monocytic cells that have phagocytosed MSCs adapt an immunoregulatory phenotype and induce regulatory T cells.
Immune Response and Optimisation:
• UC MSCs are immuno-privileged, meaning they are less susceptible to immune recognition than MSCs from bone marrow.
• The immunogenic and immunomodulatory properties of MSCs are influenced by culture conditions. We have identified treatments that may enhance some properties of MSCs e.g. pre-treatment with a cytokine cocktail or with Budesonide. Notwithstanding this, in vivo experiments show good efficacy with untreated MSCs.
• Cell density and proliferation status of MSCs have a major effect on MSC phenotype, so MSC culturing protocols merit due consideration.
Biodistribution:
• MSCs injected under the skin do not migrate to other locations.
• There is evidence from imaging analysis that dispersal of sorted MSCs may be greater from injection site than wild-type MSCs.
• Notwithstanding the remote action of MSCs, with systemic delivery of MSCs we found a greater number of cells in the liver in CCl4 injury and MDR2-KO injury mice, as compared to uninjured mice. This suggests injured livers may produce factors that attract/retain/promote survival of MSCs.
• MSCs are largely located in the lungs after administration, with an accumulation of dead MSCs in the liver 24h after infusion.
MSC Manufacture and Clinical Trial:
• Our work to date has allowed us to develop a process for MSC production from UC tissue that is consistent and aligned with regulatory requirements. NHSBT have manufactured specially selected MSCs (ORBCEL-C™, discovered by Orbsen Therapeutics) for delivery to patients in the trial.
• The MERLIN clinical trial for AIH and PSC patients opened for recruitment on 7 December 2018.
• The first patient was treated on 29 March 2019.
1.4.1.3 Exploitation and Exploitation Plans
As will be noted from the above, MERLIN has resulted in a number of key innovations. Different partners will be exploiting the results of the project in different ways in the future. Exploitation can take many forms and may include: further research building on project results; commercial exploitation (new products or services, or the furtherance of existing ones), clinical exploitation and exploitation through the development of standards, guidelines or SOPs.
Some examples of the steps being taken/planned for the further exploitation of MERLIN results are set out below.
First, it is important to note, the partners will continue to run the MERLIN clinical trial after the project ends. This will ensure the continuation of the work started in MERLIN and forms an important part of the project’s legacy.
Other ways in which MERLIN partners are exploiting and building on MERLIN include:
• UoB and ORB are involved in a new basket trial where multiple autoimmune conditions will be treated using MSCs. This will include Rheumatoid Arthritis and Crohn’s disease (aim to open this trial in 2019).
• MW-ATTC collaboration (Midlands and Wales Advanced Therapy Treatment Centre), led by MERLIN Coordinator Professor Newsome has been awarded £7.3 million funding by Innovate UK, to deliver trials to help cell or gene therapies reach the clinical market.
• MW-ATTC collaboration: ORB were awarded £1.4M to establish GMP operations and develop ORBCEL-C for the POLARISE auto-immune basket trial.
• UoB (with industry partners) has been awarded £1.1 million by UK Research & Innovation (UKRI)’s Innovate UK, to investigate patients’ experience of cell and gene therapies (PROmics study assessing the effect of novel cell therapies on patient symptoms and quality of life).
• ORB are collaborating with NHSBT to test ORBCEL-C in the Wellcome Trust REALIST clinical trial in patients with moderate to severe ARDS – with 3 patients treated in January 2019.
• EMC propose further research into long term immunomodulatory and regenerative effects, linking the MoA of MSC to morphology of MSC (modifying phenotype prior to administration to impact effect), and customising MSC therapy for particular indications, cells types and locations.
• EMC are involved in the Oxford-Aarhus-Erasmus MC consortium named MEPEP (Role of Mechynchymal Stem cells to regenerate renal graft function after ischemia reperfusion injury using normothermic ex-situ machine perfusion in pig transplantation; funded by the Lundbeck foundation, DK) where they use the gained knowledge of MERLIN to repair damaged kidneys.
• As inflammation is a well-known hallmark of cancer, UNIPD are planning to investigate the role of MSC derived EVs in mouse models of tumor progression (collaboration underway).
• UNIPD also plan to test the efficacy of MSC-EVs in the control of pathological angiogenesis characteristic for instance of eye diseases and tumours.
• SOPs and other controlled documents developed in the project will form the basis for similar developments and trials to be undertaken by NHSBT in collaboration with other academic and commercial partners. NHSBT has already started a follow-on collaboration with ORB testing ORBCEL-C in the Wellcome Trust REALIST clinical trial in patients with moderate to severe ARDS.
• BIO have further research collaborations underway and planned using their innovative 3D imaging technology (CryoVizTM) e.g. metastatic cancer imaging in projects funded by the US National Cancer Institute through the Small Business Innovative Research schemes. Potential commercial opportunities include markets in regenerative medicine, cancer immune-therapy and gene expression.
• In terms of IP protection, ORB have filed a patent application relating to CD39+ cell selection and exosome production from ORBCEL (link here). UNIPD have also filed a patent application on new anti-angiogenic extracellular vesicles derived from MSCs (link here).
• Partners have also exploited their results by delivering numerous key conference presentations and achieving several peer reviewed publications (and a PhD).
1.4.1.4 Who will benefit?
MERLIN is a truly multi-disciplinary project, encompassing the generation of new knowledge about the bio-distribution and mechanism of action of MSCs, the development of advanced technologies (3D imaging and processes for GMP-compliant manufacturing), and the delivery of an early phase clinical trial. As such, the project will have an impact on future research, manufacturing of MSCs for clinical use and cell therapies for patients with PSC and other inflammatory diseases.
Patients with PSC and AIH, who currently have no treatment options, will benefit from the project. Through delivery of the Phase 2a trial, MERLIN will provide important safety and preliminary efficacy data for a novel MSC therapy. By focusing on understanding how the therapy works to fight inflammation, we are also learning important lessons that can be used in the future to develop MSC therapies for other types of liver disease and for inflammatory conditions in general. Patients will be informed of our results through our strong links to patient groups, in particular the PSC Support group in the UK. In addition to potential benefits for individual patients and their families, improving treatments for liver disease will also benefit healthcare systems through the reduction of costly liver transplant procedures and care costs associated with chronic liver disease. From a health economics perspective, MSC therapies for immune and inflammatory diseases offer potential for significant savings in the longer term.
Researchers in liver disease, immunology and regenerative medicine will benefit from the new knowledge generated by the project and the MERLIN trial. For researchers in liver disease and in the broader field of immunology and inflammatory conditions, MERLIN is providing new data on the use of MSCs to modulate immune response and to treat inflammation. The project will also have a significant impact in the field of MSC research. We are disseminating our findings widely through publications, presentations at scientific conferences and interactions with other projects. Importantly, MERLIN is part of a larger group of sister projects with many shared partners, providing an extensive network of researchers with whom we can share ideas and results. An additional benefit derived from the project will be the advancement of 3D imaging technology, which offers researchers new, better ways to carry out bio-distribution studies.
The European cell therapy industry will also benefit from the project. We have established GMP compliant processes and quality control tests to manufacture MSCs that meet the current (and emerging) regulations for MSC therapies. In addition, our research has explored the production of optimised MSCs. The project has delivered important data and lays the groundwork for the future tailored production and use of MSCs with precise characteristics for particular applications. In order to ensure that our innovations are of use to the industry, we have focused on ensuring that our advances meet regulatory requirements.
The MERLIN consortium is committed to building on the project in future collaborations, generating long-term research value and delivering societal and economic benefits through the development of advanced MSC therapies.
1.4.2 Dissemination
A summary of the dissemination activities carried out during the MERLIN project are set out below.
1.4.2.1 Website and online
The project website http://fp7merlin.eu/ was launched in month 1 of the project and has played a central role as a public dissemination tool and means of communication. Different sections of the website are aimed at different audiences, including scientists, industry, researchers, media and the general public. The website includes the following pages:
• Home page http://fp7merlin.eu/ - a general introduction to the project.
• Project Page http://fp7merlin.eu/project/ - a more detailed look at MERLIN, with a subpage describing the research plan and a subpage for publications (both project publications and related publications of interest).
• Partners Page http://fp7merlin.eu/partners/ - details of each partner in the consortium with links to partner websites.
• Media Page http://fp7merlin.eu/media-centre/ - basic project details, press releases, contact information for the Coordinator with a subpage for promotional materials http://fp7merlin.eu/media-centre/photos-videos-and-presentations/ where project flyers, brochures, newsletters and videos can be accessed.
• Related Projects Page http://fp7merlin.eu/related-projects/ - provides links to related and complimentary research projects.
• Contact Page http://fp7merlin.eu/contact-us/ – an online form which can be used to engage with the project team.
• News Page http://fp7merlin.eu/news/ – regular updates on the project and related developments.
The website has been regularly updated over the course of the Project.
Figure 16 Screenshot of Project Homepage http://fp7merlin.eu/ Merlin Twitter Account
In addition to the website, MERLIN has an established social media presence, posting regularly on twitter (https://twitter.com/fp7merlin) and on Facebook (https://www.facebook.com/FP7MERLIN/).
1.4.2.2 Media
The project website has a dedicated media centre, which includes basic project facts, contact details for the Coordinator, the initial project press release and a factsheet from the British Liver Trust on PSC. There is also a promotional materials subpage where the project brochure, leaflet and newsletters can be accessed, together with 4 short videos featuring Merlin researchers explaining the project and the work being undertaken. The webpage makes key information about the MERLIN project easily accessible to interested journalists and to the general public
Press releases were issued at the start of the project. Press coverage included the Irish Times (“Orbsen Therapeutics to take part in €6m liver disease clinical trial’, Irish Times 22 April 2014), as well as coverage on Health Canal and My Science websites. Prof. Antonella Viola (UNIPD) was also interviewed by Italian broadcaster TG1.
A final press release has been issued, reporting that the first patient has been successfully treated in the clinical trial. The trial press release has so far been covered by the American venue, Trial Site News.
Figure 17 UoB Press Release for start of clinical trial
1.4.2.3 Awards
Our young researchers have won awards and achieved their PhDs on the basis of their excellent work in MERLIN.
1.4.2.4 Project Materials
During the course of the project we issued several project materials which were designed to engage with a broad audience.
At the start of the project we issued the project brochure and flyer setting out MERLIN’s key aims. During the life of the project we issued 2 newsletters, proving updates on progress. The MERLIN website also hosts 4 project videos that feature Merlin researchers explaining the background to their research. A public version of the report for the MERLIN Scientific Advisory Board was prepared at the end of the first and second years of the project. These reports were made available on the project website.
Partners have been provided with hard copies of project materials for distribution locally and at events and conferences (including at events hosted by the International Society for Cellular Therapy (ISCT) and International Society for Stem Cell Research).
Finally, at the end of the project we issued the MERLIN final flyer, with a high level summary of the project’s key achievements intended for a general audience. We distributed this, with cover letters, information about the project and press releases, to interested Members of the European Parliament.
All of the MERLIN project materials can be downloaded from the website: http://fp7merlin.eu/media-centre/photos-videos-and-presentations/.
Figure 18 Screenshot of some of the materials available on the project website.
1.4.2.5 Videos
The project produced a series of videos to explain our work and its value to the public.
List of Websites:
The MERLIN website is at www.fp7merlin.eu
MERLIN can be contacted via the Project Coordinator:
Prof Philip Newsome,
Centre for Liver Research
5th Floor
Institute of Biomedical Research School of Immunity and Infection
College of Medical and Dental Sciences
University of Birmingham
Edgbaston, Birmingham B15 2TT
United Kingdom
Email: p.n.newsome@bham.ac.uk
It is estimated that 29 million people in the EU have chronic liver disease, and it is the fifth most common cause of death. Most liver diseases involve inflammation that leads to liver damage. MERLIN is focused on developing a stromal cell therapy for the liver diseases primary sclerosing cholangitis (PSC) and autoimmune hepatitis (AIH). There are currently limited treatment options available for these conditions.
Researchers in MERLIN have looked at the effectiveness of Mesenchymal stromal cells (MSCs) against inflammatory liver disease in the laboratory, in order to inform and underpin a novel cell therapy. We have also explored mechanisms of action of MSCs and optimum conditions for MSC production.
Some of our key findings include:
• MSCs reduce markers of liver damage and inflammation in laboratory models of inflammatory liver disease, as well as the number of inflammatory cells in damaged areas.
• MSCs derived from both bone marrow (BM) and umbilical cord (UC) show positive effects.
• The route of infusion of UC MSC has no impact on its therapeutic effect (MSCs infused subcutaneously and intravenously were equally effective).
• The beneficial effect of MSCs on PSC is due, at least in part, to the extracellular vesicles (EVs) that MSCs secrete.
• The properties of MSCs can be influenced by the conditions under which the MSCs are cultured. Culturing protocols also merit due consideration.
• MSCs largely move to the lungs following administration, with an accumulation of dead MSCs in the liver 24h after infusion.
In MERLIN we used the results of our studies to design a clinical trial to test the safety of our MSC therapy and examine the effect of therapy on inflammation in patients with PSC and AIH. We also developed a process for MSC production from UC tissue that is consistent and aligned with regulatory requirements. NHSBT have manufactured specially selected MSCs (ORBCEL-C™, discovered by Orbsen Therapeutics) for delivery to patients in the trial. The MERLIN trial opened on 7 December 2018 and will continue (and be completed) after the project ends.
We believe that our work will set the stage for a new, MSC-based therapy for PSC and AIH, with potential application to other forms of liver disease and other inflammatory conditions. The project has delivered important data and lays the groundwork for the future tailored production and use of precisely characterised MSCs for particular applications.
In addition to potential benefits for individual patients and their families, improving treatments for liver disease will also benefit healthcare systems, through the reduction of costly liver transplant procedures and of care costs associated with chronic liver disease. From a health economics perspective, MSC therapies for immune and inflammatory diseases offer potential for significant savings in the longer term. Researchers in liver disease, immunology and regenerative medicine will benefit from the new knowledge generated by the project and the MERLIN trial. An additional benefit derived from the project will be the advancement of 3D imaging technology, which offers researchers new, better ways to carry out bio-distribution studies.
The MERLIN consortium is committed to building on the project in future collaborations, generating long-term research value and delivering societal and economic benefits through the development of advanced MSC therapies.
Project Context and Objectives:
The overall objective of MERLIN was to develop and validate (in a Phase 2a clinical trial) purified MSC as an entirely new treatment option for patients with inflammatory liver diseases PSC and AIH, in parallel with enhancing knowledge about the operation of MSCs.
Background
The prevalence of chronic liver disease has been estimated to be 6% (or 29 million people) in the EU (The European Liver Patients Association. 2005), with mortality rates estimated at 14.3 per 100.000 in the EU-25 in 2004, making it the 5th commonest cause of death (the EU Statistical Year Book, 2007). Most liver diseases have a significant inflammatory component that underpins liver damage & fibrogenesis, yet current therapies have limited effectiveness. Primary Sclerosing Cholangitis (PSC) is an example of chronic immune mediated liver injury, with a natural history of progression to liver failure. The incidence and prevalence rates for PSC vary from 0 to 1.3 per 100,000 inhabitants/year and 0 to 16.2 per 100,000 inhabitants, respectively. Liver transplant is ultimately the only viable treatment for many people with PSC.
Our challenge in MERLIN was to develop a therapy that alleviated the ongoing inflammation that drives fibrosis in liver injury. Using PSC and AIH as our paradigm reflected the inflammatory and immune mediated nature of these conditions, alongside the huge unmet need for therapy. We aimed to harness laboratory and clinical studies to effectively, and in a timely manner, answer our questions regarding optimised MSC therapy and how such therapy might be applied in the context of PSC and AIH.
MERLIN Key Objectives
As noted above, the main objective in MERLIN was to develop a Phase 2a clinical trial administering purified MSC as an entirely new treatment for PSC and AIH patients, in tandem with developing new knowledge about MSCs.
Some more specific and detailed objectives of the project described in the original programme of work are set out below.
Translational Aims (related to the MERLIN clinical trial)
The aims below were devised at the outset of the project to help us deliver the proposed clinical trial.
• To compare the safety and efficacy of ORB1+ and ORB1- MSC to PA-MSC in refined models of PSC – this aim was designed to help us select the most appropriate MSC-based product for use in the MERLIN trial.
• To ensure EU ATMP compliance and prepare an application for a Phase 2a Clinical Trial of ORB1 therapy in PSC – this aim involved the generation of an Investigational Medicinal Product Dossier (IMPD) Investigator Brochure (IB) and Clinical Trial Protocol for the MERLIN Clinical Trial Application to the Medicines Health and Regulatory Authority (MHRA) in the UK.
• To carry out a Phase 2a Clinical Trial of ORB1 therapy in PSC – this aim related to the validation of the selected MSC-based therapy in the MERLIN clinical trial.
• To establish the in vivo immunological footprint of MSC after infusion into pre-clinical models of PSC and also in clinical trial patients – this aim set out practical tests to be applied in order assess the impact of the therapy administered.
Development Aims (increasing knowledge about MSCs)
The following aims were identified at the start of the project in order to guide our studies into MSCs and help us develop a deeper understanding of MSC operation.
• To elucidate the mechanism of action of MSC in ameliorating liver inflammation -this aim required us to explore the role and mechanisms of action (MoA) by which MSC simultaneously alleviate liver inflammation and repair tissue damage.
• To improve the functional longevity of MSC – this aim required an assessment of culture protocols for MSCs, to determine at what point in the culture process their efficacy may be compromised.
• To reduce the immunogenicity of MSC – this aim focused on exploring how to modulate the potential susceptibility of MSC to immune cell recognition, in an effort to improve survival of MSCs post infusion.
• To improve the homing/localisation of MSC – this aim required us to consider the homing and biodistribution of MSCs post infusion and the most effective ways to administer MSCs.
• To generate the optimal MSC – this aim required the study of conditions which might optimise the operation of MSCs.
Legacy Aims (To maximise the long-term research and collaboration value of MERLIN).
Specific aims in relation to MERLIN’s long-term legacy were clearly articulated in the MERLIN programme of work:
• To agree a Memorandum of Understanding between the partners to guide long term collaboration with potential to expand and add new members.
• To agree standard operating protocols for isolation, culture, storage, transport and use of MSC.
• To publish training and best practice materials for the wider research community.
• To develop training opportunities across the consortium.
Conclusion
We put the aims and objectives outlined above into practical application as we embarked on the MERLIN work programme. We explored the Development Aims described and discovered new knowledge about MSCs. MERLIN partner BIO succeeded in further developing their 3D imaging technology (CryoVizTM) for high-throughput imaging and analysis. Different elements of our research emerged as more or less important as our studies generated novel results and directed our focus.
We applied the new information learned in the design of the MERLIN clinical trial, which is now underway. NHSBT have manufactured specially selected MSCs (ORBCEL-C™, discovered by Orbsen Therapeutics) for delivery to patients in the trial.
Details of the scientific work undertaken and key results achieved are set out in section 1.3 below.
Project Results:
Introduction
Mesenchymal stromal cells (MSCs) represent a promising therapeutic approach in many diseases, including inflammatory liver disease, in view of their potent immunomodulatory properties. Researchers in MERLIN have looked at the effectiveness of MSCs against inflammatory liver disease in laboratory models, in order to help develop a novel cell therapy for PSC and AIH. A clinical trial was designed based on the outcomes of our studies in the laboratory. The trial is now underway, looking at the safety of the MSC therapy and examining the effect of therapy on inflammation in patients with PSC and AIH. NHSBT have manufactured specially selected MSCs (ORBCEL-C™, discovered by Orbsen Therapeutics) for delivery to patients in the trial. This will set the stage for a potential new, MSC-based therapy in the future. In addition, MERLIN is generating new knowledge that is more widely applicable to MSC therapy in general, informing the development of therapies for other forms of liver disease and other inflammatory conditions.
The main research undertaken and results achieved are described below under the following headings:
• WP1 Pre-clinical efficacy testing of MSC.
• WP2 Functionality of endogenous MSC.
• WP3 Immunogenicity and immunological footprint of MSC in PSC.
• WP4 MoA of MSC in models of liver damage in vivo.
• WP5 Genetic, biological & pharmacological enhancement of human ORB1+MSC re liver inflammation treatment.
• WP6 Advanced tools for bio-distribution of MSC.
• WP7 Generation of Investigational Medicinal Product Dossier and Manufacture of Clinical Grade ORB1+/- MS
• WP8 Phase 2a clinical trial of ORB1+/- MSC in patients with PSC
In addition, the data management and samples management work carried out in WP9, to support the S&T core, are briefly described.
1.3.1 WP1 Pre-clinical efficacy testing of MSC
1.3.1.1 WP1 Introduction
Encouraging results obtained from pre-clinical studies have highlighted the immunomodulatory and clinical potential of mesenchymal stromal cells (MSCs) Consequently, many autologous and allogenic MSC based clinical trials are being undertaken for a number of indications including graft versus host disease (GVHD) and acute and chronic liver disorders. Nevertheless, transplantation of third party MSCs has been shown to be feasible and safe with good tolerability profiles in clinical studies. There are still uncertainties about their long-term therapeutic efficacy. Although some clinical studies have undoubtedly demonstrated clinical efficacy of MSC, most of the studies often yield mixed results, suggesting a significant variability in the immunomodulatory potency of MSC. It is now accepted that the therapeutic potential of MSCs mainly relies on their short-term paracrine ability to reduce inflammation, inhibit immune responses, and produce trophic factors. An important requirement for the clinical usage of MSC would be the production of clinical grade cells with better treatment efficiency. Therefore, in this reporting period in WP1 we set out to investigate whether pre-treatment of ORB+ UC-MSC with either IFN-γ+TGF-β +retinoic acid (RA) or the glucocorticoid steroid budesonide would enhance their immunomodulatory and therapeutic efficacy in the Mdr2Ko/FVB pre-clinical murine model of inflammation mediated liver injury. WP3 and WP5 have previously demonstrated that priming of ORB+ UC-MSC with EMC-cocktail or budesonide enhanced their immunomodulatory potential in vitro. Also, their mechanism of action will be elucidated using mono- and co-culture of MSC with hepatic sinusoidal endothelial cells (HSEC) and biliary epithelial cells (BEC) in vitro.
1.3.1.2 WP1 Key Work Undertaken and Results Achieved
The overall objective of MERLIN is to bring optimised second-generation MSC therapy to the clinic to treat patients with Primary Sclerosing Cholangitis (PSC).
Cellular senescence of hepatocytes, cholangiocytes, stellate cells and immune cells has been implicated in chronic liver disease progression. We have showed that hydrogen oxide can induce senescence (growth arrest) in human biliary epithelial cells. Furthermore, treatment with MSC or MSC conditioned medium reduced biliary epithelial cell senescence adding further to their beneficial effects. In addition, MSC suppressed the proliferation/activation of injurious inflammatory cells (lymphocytes) isolated from both the peripheral blood of patients with PSC and the explanted liver of patients with PSC who have undergone liver transplant. These data confirm the beneficial effects of ORB+ MSC on human effectors of inflammatory liver disease.
Our pre-clinical studies demonstrated that ORB+ MSC from bone marrow and umbilical cord attenuated hepatic inflammation in Mdr2KO/FVB models of biliary injury. Notably, MSC infusion improved liver injury as evidenced by a reduction in serum alanine aminotransferase (ALT) and bile acid (BA) levels, hepatic pro-inflammatory immune infiltrates and the generation of T regulatory cells and anti-inflammatory macrophages. We demonstrated this effect held true for other models of liver injury including Ova-Bil and CCl4.
Our in vitro data demonstrated that MSC could exert their inhibitory effects locally and/or remotely to down regulate leukocyte recruitment. We also described the involvement of IDO in this phenomenon since blocking this enzyme with 1-MT removed the inhibitory effect of not only MSC but also conditioned medium collected from unstimulated/stimulated MSC. This could in turn reduce the expression of VCAM-1 and ICAM-1. Building on this work we were able to demonstrate that MSC could exert similar beneficial effects in vivo after subcutaneous administration suggesting their ability to exert their effects remotely without a requirement to home to the liver.
We sought to establish if we could further improve the effectiveness of MSC by pre-treating them with cytokines. Having tried multiple cytokine combinations we were able to see changes in gene expression but no major differences in terms of their efficacy in vivo in models of liver injury. Therefore, this study suggests that good clinical grade unstimulated ORB+ UC-MSC may substantially contribute to promising clinical outcomes in an appropriate physiological pro-inflammatory micro-environment without any pre-treatments.
1.3.1.3 WP1 Conclusion
This work package has been successful in demonstrating the effectiveness of MSC in reducing liver injury in a number of different models as well as providing evidence of efficacy with human cells. It also provides insights as to the mechanism of action which will be useful in future studies.
1.3.2 WP2 Functionality of endogenous MSC.
1.3.2.1 WP2 Introduction
This work package, WP2, aims to elucidate the functionality of endogenous MSCs during liver injury by significantly depleting their numbers and subsequently monitoring the development of liver injury in specific mouse models. In 2010, Kraman and colleagues developed a conditional knock-out murine model, which allows the depletion of stromal cells, including bone marrow MSCs. These transgenic mice are biologically engineered to express the human diphtheria toxin receptor (DTR) and Luciferase (Luc)2 in cells which naturally express fibroblast activation protein (FAP), a surface protein marker identified on a range of stromal cells, including human and mouse MSCs (Bae et al., 2008; Tran et al., 2013). Administration of diphtheria toxin (DTx) to these mice results in the depletion of any cell expressing FAP via the engagement of the DTR protein. This FAP-DTR murine model should allow us to determine the impact of stromal cell depletion, including populations of MSCs, during liver injury. A limitation of this model is that a range of other non-MSC FAP+ stromal cell populations, identified in muscle, pancreas and lymph node tissues, are also targeted by the DTx treatment, making it difficult to directly attribute any effects observed to the depletion of MSCs. Nevertheless, this initial study could potentially inform future more in depth studies.
1.3.2.2 WP2 Key Work Undertaken and Results Achieved
We have demonstrated that the expression of the FAP protein was significantly increased at the gene and protein level in chronically diseased human liver tissue when compared to normal liver, as confirmed via qPCR and western blotting. Also, we showed the tissue distribution of FAP via immunohistochemical staining is largely restricted to portal areas and fibrotic septa in chronically diseased tissues. Furthermore, we showed that FAP protein expression strongly correlated with the extent of fibrosis in the diseased tissues, as measured by picrosirius red analysis. This was also confirmed at the mRNA level, with FAP mRNA expression exhibiting a strong positive correlation with collagen mRNA expression (Coll IAI). Additionally, in the previous periods, we started to characterise the expression of FAP in murine liver tissues, with initial experiments demonstrating an increase in FAP mRNA expression in the biliary injury model, MDR2-/-, but no significant changes were observed in acute and chronic carbon tetrachloride (CCl4) models of injury.
Finally, we were able to optimise the murine model which utilised diphtheria toxin (DTx)-mediated depletion of FAP+ stromal cells followed by acute CCl4-driven liver injury. However, it became apparent that the depletion of FAP+ stromal cells significantly affected a number of tissues due to the wide expression of FAP in tissues, such as muscle, pancreas and lymph nodes, making it difficult to directly attribute any effects observed to the depletion of MSCs.
Nevertheless, our results suggest that endogenous populations of stromal cells are not simply bystanders during the administration of exogenous MSCs and could potentially play a role in the pathology of both acute and chronic liver injuries. However, it is unclear from the current data whether that role is beneficial or in fact detrimental. Additionally, these data suggest that the effects of depletion are very much dependent on the context of injury type (acute or chronic) and stage (active injury or resolution) and it is evident that further investigation, beyond the scope of MERLIN, is warranted.
1.3.2.3 WP2 Conclusion
We have comprehensively characterised the expression of FAP in murine and human liver injury which will be of significant value to the scientific community. In so doing it has become clear that FAP expression is not limited to just MSC and thus interpretation of knock-down of FAP is challenging. Our data set the scene for further studies that will need to refine this model further if the effect of MSC depletion alone is to be understood.
1.3.3 WP3 Immunogenicity and immunological footprint of MSC in PSC.
1.3.3.1 WP3 Introduction
WP3 developed around 3 key questions concerned with optimisation, mechanisms of action and monitoring of MSC treatment for immunoregulatory therapy. The first question was whether pre-treatment of MSC in vitro could optimise their immunoregulatory efficacy and reduce their immunogenicity. This question was addressed by treating MSC with a variety of factors and determining the effect on MSC immunophenotype, gene expression profile, immunoregulatory function and susceptibility for cytotoxic lysis. We furthermore examined the effect of prolonged culture on MSC phenotype and function and we performed whole genome methylation profiling to examine whether culture-induced changes would leave an epigenetic imprint in MSC.
The second question in WP3 was to elucidate the mechanism of action of MSC therapy. While it was previously thought that the effect of MSC is dependent on the persistence of MSC after infusion, we and others determined that MSC are short lived after infusion. This raises the question of how MSC modulate the immune system after infusion, which we addressed by tracking experiments and measurement of immune parameters after MSC infusion. We also examined whether MSC with optimised immune properties would show a different biodistribution and survival pattern after infusion.
The third question was whether MSC infusion would leave an immunological footprint in PSC patients. To examine this we performed in depth analysis of the immunological footprint in different groups of inflammatory liver disease patients and compared them to healthy controls. We determined immune cell subset distribution and performed serum analyses. These experiments served to setup methodology for measurements of samples from the Merlin clinical trial and provided a background pattern of immune parameters in PSC patients.
1.3.3.2 WP3 Key Work Undertaken and Results Achieved
Experiments were carried out in vitro to treat umbilical cord-derived MSC (ucMSC) with a variety of cytokines, growth factors, vitamins and prostaglandins and examine their immunophenotype, gene expression profile and functionality in T cell suppression assays. Of all tested factors, IFNg was the most potent factor in upregulating the expression of key molecules in MSC involved in their immunoregulatory effect, including IDO and PD-L1, and in enhancing the capacity of MSC to suppress T cell proliferation and IFNg production. At the same time, IFNg containing pre-treatments increased the expression of HLA class I and II molecules on MSC, impacting the susceptibility of MSC for cytotoxic attack. To examine whether pre-treatment would modify the epigenome of MSC, whole genome methylation analysis was performed. Only minor effects on the DNA methylation profile of MSC were detected, indicating that the phenotypical changes induced by pre-treatment were transient. Pre-treated MSC were administered via tail vein injection in a CCl4 inflammatory liver disease mouse model. The differentially treated MSC showed no differences in biodistribution and accumulated in the lungs. There was also no difference in MSC survival and there was no therapeutic effect of control MSC or pre-treated MSC in this model in our hands. Possibly the lack of MSC distribution to the liver prevented a potent effect of MSC on liver inflammation. Therefore we set up an ex vivo liver tissue slice model. In this model MSC treated with a combination of IFNg, TGFb and retinoid acid were most potent in reducing the expression of inflammatory factors after challenging the tissue with LPS (Figure 1).
In addition to treating MSC with growth factors and cytokines, we examined the effect of prolonged culture on the phenotype and function of MSC. Culturing MSC higher passage numbers (from 4-12) reduced their proliferation and their capacity to inhibit T cell proliferation. It was also observed that within a passage MSC showed large differences in DNA methylation patterns from time of seeding until confluency. This was most likely associated with cell density.
Figure 1 Effect of differentially pre-treated ucMSC on ex vivo liver slices. Gene expression levels and cytokine/chemokine levels were measured in liver slices treated with LPS and supernatant, respectively. (a)(b)(c) Both gene expression levels and cytokine secretion levels were measured for MCP-1, TNFα and IL6. In addition, (d) gene expression levels of IP10 and also (e) CXCL1 was measured in the medium were measured. n=6 * p<0.05.
The mechanisms of action of MSC therapy were explored by examining the fate of infused MSC. As explained above, intravenously infused MSC accumulate in the lungs. Double labeling of MSC with Qtracker605 beads, which would disappear from the cells after cell death, and Hoechst, which would be preserved even after cell death, allowed tracking of infused MSC even after death. It was found that directly after infusion the majority of the cells is alive and present in the lungs. After 24h most cells have died and translocate partly to the liver. After 72h no living or dead MSC were detected anymore (Figure 2). The presence of MSC in the lungs was accompanied with elevated numbers of monocytes and neutrophils in lung tissue, which phagocytosed the administered MSC. Monocytes that phagocytosed MSC were subsequently detected in the blood and in liver tissue. Upon phagocytosis of MSC, monocytes adapted a regulatory phenotype and adapted the capacity to induce regulatory T cells. .
Figure 2 Quantitative detection of double-labeled MSC after intravenous infusion in whole mice, and specifically in lungs and liver. The Qtracker signal detects living MSC, whereas the Hoechst signal also detects dead MSC.
To examine the immunological footprint of PSC patients and other inflammatory liver disease patients, blood samples were analysed for immune cell subset frequencies, profiles of plasma cytokines and growth factors and concentration and size of nanoparticles in plasma. We determined naïve, central memory, effector memory and late effector memory CD4 and CD8 T cell frequencies, B cells, NK cells and different types of monocytes and dendritic cells. Distinct immune cell profiles were observed for different groups of inflammatory liver disease patients. PSC patients for instance showed increased percentages of CD19+ B cells compared to other inflammatory liver disease patients and healthy controls (Figure 3). Furthermore, on the basis of plasma cytokine and growth factor profiles and nanoparticle levels distinctions between different patient groups and healthy controls could be made. The deviated immunological footprint of PSC patients and other inflammatory liver disease patients compared to healthy controls indicates that immune profiling of patients in MSC trials may reveal efficacy and safety information.
Figure 3 Frequencies of CD19+ B cells, CD16+CD56+ NK cells and CD14+ monocytes in liver disease patients and healthy donors.
1.3.3.3 WP3 Conclusion
WP3 revealed novel mechanisms of action of MSC therapy. MSC end up in lung tissue after infusion, where they interact with innate immune cells that subsequently adapt a regulatory phenotype and migrate to other sites of the body. Methodology was set up and tested to track MSC after administration and to determine effects of MSC therapy. The immunological footprint of inflammatory liver disease patients was determined, which will serve as a baseline for the Merlin clinical trial.
1.3.4 WP4 MoA of MSC in models of liver damage in vivo.
1.3.4.1 WP4 Introduction
The overall goal of MERLIN is to develop and validate a new treatment option for patients with Primary Sclerosing Cholangitis (PSC) and enhance the functionality of Mesenchymal Stromal Cells (MSC). MSC represent a promising therapeutic approach in many diseases, including inflammatory liver disease, thanks to their potent immunomodulatory properties . MSC inhibit in fact T cell activation induced by an anti-CD3/CD28 antibody stimulus, mitogens, and alloantigens; they also inhibit NK cell activation, as well as B cell terminal differentiation, and dendritic cell maturation and functionality. However the molecular mechanisms responsible for the anti-inflammatory effects of MSC in liver disease are still unknown. Much of the pre-clinical work in liver injury showed that MSC reduce oxidative stress and cellular infiltrate in the liver but it has not been yet defined the mechanism by which this was achieved.
The specific objectives of WP4 were:
1. to identify the environment/locational requirements for MSC-mediated immunomodulation in vivo
2. to identify the molecule/s responsible for the anti-inflammatory effects of MSC in vivo
3. to identify the action of MSC on the hepatic vasculature in suppressing inflammation
1.3.4.2 WP4 Key Work Undertaken and Results Achieved
In order to identify the environment/locational requirements for MSC-mediated immunomodulation in vivo we exploited MSC encapsulation that provides conclusive data in support of either the endocrine or the paracrine/contact signalling required for MSC immunomodulation . Thus, we injected alginate-encapsulated human CD362+ MSCs (E-MSCs) in a mouse model of local inflammation induced by s.c. injection of CFA/OVA . Our results clearly showed that BM CD362+MSCs exert their immunomodulatory properties in vivo, even better than unsorted MSCs, but encapsulation may be detrimental in the case of human, sorted MSCs. (Figure 4)
To formally demonstrate that s.c. injected WT or CD362+MSCs do not leave the injection site, in collaboration with Bioinvision, we performed whole-mouse cryo-imaging analysis of mice immunized with CFA/OVA and injected with fluorescently labelled hMSCs, confirming that neither WT nor CD362+MSCs migrated away from the site of injection and are able to dampen inflammation through the release of soluble mediators.
In an effort to understand the molecular mechanisms responsible for the endocrine effects of MSC, we exploited a standardised workflows previously developed in our laboratories for shotgun proteomic characterization of the mouse MSC secretome . Our results on MSC secretome indicated that, upon activation by inflammatory cytokines, MSC upregulate the expression of several proteins potentially affecting angiogenesis and inflammation through multiple pathways. Interestingly, when we compared the secretomes of human and mouse MSCs, we found that only 16 proteins are upregulated in both cell types and 11 of them modulate angiogenesis directly or indirectly, thus supporting the idea that the endothelium is a specific target of MSC during inflammation. Among the molecules upregulated upon cytokine stimulation, we focused our attention on the tissue inhibitors of metalloproteinases 1 (TIMP1). Thus, we assessed the role of TIMP-1 secreted by MSC in the CFA/OVA mouse model of local inflammation, demonstrating that MSC inhibit high endothelial cell activation and lymphocyte homing to lymph nodes by releasing TIMP-1 and further identifying the endothelium as a specific and novel target of MSC-based therapy. (Figure 5). Therefore, we proposed that pre-activation with pro-inflammatory cytokines strengthened the anti-angiogenic effects of MSC, thus supporting our hypothesis that, during an inflammatory response, MSC target angiogenesis and thus dampen the inflammatory response. Additionally, in agreement with data obtained with mouse MSCs, soluble factors secreted by human MSCs affected the tubulogenesis of HUVEC cells . However, in the case of human cells, MSC-conditioned media (CM) was able to inhibit tube formation even when MSC had not been primed by cytokines. Of note, in the case of human MSC, TIMP-1 blockade restored the formation of the HUVEC endothelial network in the presence of either unstimulated (unstim) or stimulated (stim) human MSC-CM, thus pinpointing TIMP-1 as a crucial MSC-derived modulator of angiogenesis also in human models. (Figure 6)
To investigate the role of MSC in modulating angiogenesis in inflamed livers, we performed tube formation assays of human sinusoidal endothelial cells (HSECs) from PSC patients in the presence of CM from unstimulated or cytokine-stimulated bone marrow human MSCs. Of note, both unstimulated and stimulated MSC-CM induced a significant reduction in tube formation of HSECs in comparison with control medium (aMEM). Further, we assessed the contribution of TIMP-1 in inhibiting angiogenesis of HSECs in the presence of human MSC-CM. To this purpose we performed tube formation assays of HSECs by adding anti-TIMP-1 blocking antibody to CM from both stimulated and unstimulated bone marrow hMSCs, using the appropriate isotype as control (Figure 7). Our results clearly show that TIMP-1 plays a fundamental role in the inhibition of angiogenesis also of liver sinusoidal endothelial cells.
Further, we investigated the contribution of TIMP-1 in controlling liver inflammation, by overexpressing TIMP-1 in MDR2 -/- mice by AAV9-mediated gene transfer and analyzed liver immune cell infiltration. AAV-TIMP-1 injection resulted in reduced liver infiltration of myeloid cells and did not affect Kupffer cells population. Moreover, TIMP-1 overexpression reduced the amount of neutrophils and infiltrating macrophages and increased the amount of restorative macrophages in comparison with control untreated MDR2 -/- mice. Further, AAV-TIMP-1 treated mice showed lower levels of bile acids, ALT and ALP than untreated MDR2 -/- mice (Figure 8), suggesting a beneficial effect of TIMP-1 overexpression in PSC mouse model. However, the AAV-LacZ injected mice used as AAV-infection control collectively showed less injury than untreated MDR2 -/- mice, suggesting that the observed effects were not specifically dependent on AAV9-TIMP-1. We also found that high levels of TIMP-1 negatively correlate with Treg accumulation in injured livers of MDR2-/- mice. Thus, we blocked TIMP-1 by administering an anti-TIMP-1 blocking antibody in MDR2 -/- mice. Nonetheless, TIMP-1 blockade did not result in amelioration of clinical score of liver injury, thus indicating that, although TIMP-1 plays an important role in dampening liver immunosuppression, other factors might contribute to liver injury in MDR2-/- mouse model (Figure 9).
Finally, we assessed the effects of extracellular vesicles (EVs) derived from human BM MSC in our PSC mouse model, finding that EVs from MSC might improve the clinical outcome of MDR2-/-mice.
Figure 4 Effect of hMSC on OVA-specific T cell proliferation. The experimental protocol was designed to investigate the influence of hMSC upon the activation of OT62 (OVA6specific) T cells: OT2 CD45.2 CFSE6labelled cells (1x106) were transferred into CD45.1 C57 mice. After 24 hrs mice were immunized with OVA peptide (100 ng) in complete Freund s adjuvant (CFA) by s.c. injection into the footpad. On day 2 a group of animals received 500.000 hMSC wt injected s.c. in the back another group received 500.000 hMSC S+ s.c. and the last group received 500.000 encapsulated (E6)hMSC S+ s.c.. Immunized mice were sacrificed at day 4 the popliteal draining LNs (dLNs) were collected and cells were analysed by flow cytometry.
Figure 5 TIMP-1 mediates the anti-angiogenic effect of MSC-CM in vitro and the anti6inflammatory effect of MSC in vivo.SVEC4-10 network formation in matrigel in the presence of MSC6CM or unst MSC6CM and anti mTIMP1 blocking antibody. A) Representative images at 6 hrs (left). Segment length quantification (right). Data are expressed as mean ± SEM (*p<0.05 **p < 0.01; One way Anova). B) Diagram of the experimental protocol designed to investigate the contribution of TIMP61 to the anti6inflammatory effects of MSC. Mice were immunized in the dorsal back with CFA/OVA on day 0 and, on day 1, 3 groups of animals received subcutaneous injection of 106 MSC in the lumbar back. 18 hours after MSC transplantation goat polyclonal anti-TIMP-1 IgG or isotype-matched goat IgG was i.v. administrated. On day 4 brachial LNs were collected, processed and analysed by flow cytometry; C) Graph showing the absolute number ofCD456CD31+ cells per single LN expressed as normalized percentage on CFA/OVA (3independent experiments, n=28 dLN). D) Diagram of the experimental protocol designed to over6express TIMP61 in immunized mice. One day after AAV9 administration (day 0),mice were immunized with CFA/OVA. Brachial draining LNs (dLNs) were collected 4days after immunization and digested for FACS analysis. E6G) Graph showing LNcellularity, absolute count of CD45+ and CD456CD31+ cells per single LN expressed as normalised percentage on CFA/OVA. In E6G error bars represent standard error, n= 6 mice for each condition (*p < 0.05; **p < 0.01; t-test
Figure 6: Timp-1 blocking reverts the anti-angiogenic effect of mouse and human MSC conditioned media. MSC-derived TIMP-1 concentration in (A) mouse and (B) human unstimulated or stimulated MSC conditioned medium was measured with ELISA. Data are expressed as mean ± SEM (*p<0.05 parametric T Test). Tube formation assay was performed in the presence of (C) mouse or (D) human TIMP-1 blocking antibody. Representative images of (C) SVEC4-10 cell line or (B) Huvec cells are taken with a phase contrast inverted microscope at 4× objective magnifications. Graphs show the quantification of the tube segment length measured with ImageJ Angiogenesis Analyzer. Data are expressed as mean ± SEM (*p<0.05 **p<0.01; One way Anova).
Figure 3
Figure 7
Figure 8
Figure 9 TIMP-1 modulates regulatory T cells accumulation by regulating TGF beta activation in MDR2-/- mouse livers. The role of TIMP-1 in modulating liver inflammation was assessed by using anti-TIMP-1 blocking antibody. Experimental schedule is represented in the panel (A). 6-week-old MDR2-/- mice were i.p injected twice a week with anti-TIMP-1 blocking antibody, using isotype as control. Mice were sacrificed 7 days after the first injection and liver and blood were harvested. Livers were mechanically and enzymatically (Collagenase II and DNAse I 0,2mg/ml at 37°C for 1hour) digested for FACS analysis of immune cell infiltration. Graphs show regulatory T cells (CD45+ CD3+ CD4+ CD25+ FoxP3 + within the CD45+ leukocytes) (B) and amount of active TGF beta, measured by ELISA (C). In the panel D, IHC for CD31, marker of endothelial cells, show the altered liver tissue in MDR2-/- mice. Blood was processed to obtain serum and analysed for the quantification of Bile Acid (E), ALT (F), and ALP (G). n= 4 mice per group, data are expressed as mean ± SEM (*p<0.05 **p<0.01 One way Anova).
1.3.4.3 WP4 Conclusion
Exploiting a mouse model of local inflammation, we demonstrated that MSC inhibit high endothelial cell activation and lymphocyte homing to lymph nodes. In order to clarify MSC mechanism of action, we performed a high-throughput analysis of MSC secretomes that provided us a plethora of pharmacologically targetable molecules for the treatment of inflammation. Among the upregulated proteins, we focused our attention of TIMP-1, providing evidence that TIMP-1 secreted by MSC plays a fundamental role in the inhibition of angiogenesis in vitro and in vivo, inhibiting T-cell homing into inflamed lymph nodes.
Importantly, we found a common antiangiogenic signature in the secretome from both mouse and human MSC stimulated with proinflammatory cytokines. In agreement with data obtained with mouse MSCs, we also showed that soluble factors secreted by human MSCs affect the ability of HUVEC cells to form tubes. Interestingly, in the case of human cells, MSC-CM was able to inhibit tube formation even when MSC had not been primed by cytokines. However, pre-activation with pro-inflammatory cytokines strengthened the anti-angiogenic effects of MSC-CM, thus supporting our hypothesis that, during an inflammatory response, licensed MSC target angiogenesis and thus dampen the inflammatory response.
Collectively, our results clearly show that TIMP-1 plays a fundamental role in the MSC-mediated inhibition of angiogenesis in vitro and in vivo model of local inflammation. We confirmed our findings by exploiting mouse and human MSC, as well as both mouse ad human endothelial cells. Importantly, we found that TIMP-1 plays a fundamental role in the inhibition of angiogenesis also of liver sinusoidal endothelial cells from PSC patients. However, TIMP-1 AAV9 mediated gene transfer in MDR2-/- mice resulted in only a partial beneficial effect. Of note, this effect seemed also to be independent from TIMP-1 and likely associated with AAV vectors injections.
Nevertheless, although high levels of TIMP-1 negatively correlate with Treg accumulation in injured livers of MDR2-/- mice, TIMP-1 blockade did not result in amelioration of clinical score of liver injury in MDR2-/- mice, thus indicating that other factors might contribute to liver injury in this mouse model. Finally, we found that extracellular vesicles (EVs) derived from human BM MSC might improve the clinical outcome of MDR2-/-mice.
1.3.5 WP5 Genetic, biological & pharmacological enhancement of human ORB1+MSC re liver inflammation treatment.
1.3.5.1 WP5 Introduction
The overall objectives of WP5 were:
• To isolate, expand, and cryopreserve CD362+MSC and PA-MSC from human bone marrow and umbilical cord tissues under GLP conditions.
• To provide GLP-compliant MERLIN SOPs and BMRs to partners to define, govern and record the thawing and expansion of cryopreserved human ORB1+MSC and PA-MSC in each WP laboratory.
• To assess the impact of thawing/washing of cryopreserved stromal cells upon efficacy in CCl4-induced model of liver Injury.
• To assess the impact of prolonged cell culture passage upon MSC efficacy in vitro and in CCl4-induced model of liver Injury.
• To determine the impact of ORB1-overexpression upon MSC efficacy in CCl4-induced model of liver injury.
• To explore unmodified and ‘optimised’ ORB1+ MSC in CCl4-induced model of liver injury.
1.3.5.2 WP5 Key Work Undertaken and Results Achieved
In Period 1, WP5 focussed on the manufacture of cryopreserved CD362+ MSC and PA-MSC from human bone marrow and umbilical cord tissue - and delivered cells to partners in WPs 1, 3, 4, 5 and 6 for testing. >200 vials of cryopreserved CD362+, and PA-MSC from human marrow and umbilical cord were delivered to MERLIN partners for pre-clinical experiments.
In addition, studies were completed that show no impact of cryopreservation/thawing on CD362+ MSC efficacy in the CCl4-induced model of liver injury in the C57Bl6 mouse.
The impact of the glucocorticoid Budesonide on the efficacy of ORB1+ MSC in suppressing T lymphocyte proliferation was also assessed. Pre-treatment of CD362+ MSC with Budesonide stimulates expression of the CD362 protein, GILZ mRNA and TIMP1 RNA. Notably, Budesonide-treated ORB1+ MSC significantly improved MSC-mediated suppression of T lymphocyte proliferation.
In total ORB delivered over 400 vials of cryopreserved stromal cells to MERLIN partners over the course of Period 1 and Period 2. ORB explored novel activators of ORBCEL-C and focused on developing Budesonide as an activator of ORBCEL-C.
Researchers have worked on enhancing the manufacturing of the ORBCEL-C product using the MACSQuant Tyto to select near 100% pure stromal cells and expanding ORBCEL-C on the closed automated Quantum Bioreactor and this led to a collaboration between ORB and NHBST to test Quantum expanded ORBCEL-C in the REALIST clinical trial (https://clinicaltrials.gov/ct2/show/NCT03042143)
ORB have explored the potency of ORBCEL-C with the CD4 T cell proliferation assay and VCAM1-specific Endothelial cell activation assay and discovered a biomarker (ORB2) that allows ORB to select effective cord tissue donors for ORBCEL-C Manufacture.
ORB collaborated extensively with NHSBT and UOB over the entire MERLIN project to transfer the ORBCEL-C Manufacturing process to NHSBT for the MERLIN Phase 2a clinical trial. This has allowed the MERLIN trial to progress to treat its PSC patient in March 2019.
https://www.prnewswire.co.uk/news-releases/clinical-trial-investigates-orbsen-therapeutics-orbcel-c-tm-immunotherapy-as-treatment-for-two-autoimmune-liver-diseases-820613487.html
Finally, ORB have shown that ORBCEL-C derived extracellular vesicles express CD362, CD39 and CD73 and retain intrinsic ATPase activity. With a view to understanding the consistency of the ORBCEL-C product, ORBCEL-C derived from 7 donors in 8 series of experiments was tested in a rat model of lung inflammation and showed consistent function of the ORBCEL-C Drug Product.
1.3.5.3 WP5 Conclusion
MERLIN WP5 enabled ORB to develop the ORBCEL-C selection and GMP manufacturing process from the pre-existing ORBCEL-M marrow CD362-selection & expansion process developed in the REDDSTAR program. ORB then manufactured and distributed >400 vials of ORBCEL-C for pre-clinical testing, optimisation and in vitro evaluation in WP1, WP3, WP4, WP5 and WP6. ORB have tech transferred the ORBCEL-C Manufacturing process to NHSBT in WP7 to allow clinical manufacture of ORBCEL-C for the MERLIN clinical trial in WP8. ORB have established the 2 year shelf life of the ORBCEL-C Drug Product. ORB have since established GMP manufacturing operations and are collaborating with UOB to design a Phase 2b MERLIN2 Adaptive Licensing in patients with autoimmune liver disease.
1.3.6 WP6 Advanced tools for bio-distribution of MSC.
1.3.6.1 WP6 Introduction
BioInVision, based in Cleveland, Ohio, USA, offers imaging instrumentation and methodologies critical to preclinical studies. The unique CryoVizTM instrument, utilizing the patented cryo-imaging technology, allows microscopical anatomical and molecular fluorescence imaging of laboratory small animals such as a mouse or organs excised from them with single-cell sensitivity. With its sub-ten-micron-scale imaging, cryo-imaging allows one to detect even single stem or cancer cells anywhere in a mouse. The technology is also offered as a service and is targeted to a variety of biomedical applications including stem cell homing and biodistribution, cancer metastasis, imaging agents, drug discovery and delivery, tissue engineering, mouse phenotyping etc. BioInVision’s cryo-imaging technology was a good match for the MERLIN project which focused on pre-clinical and clinical studies to investigate specifically the use of MSCs as a therapy for Primary Sclerosing Cholangitis (PSC), a type of liver disease for which there is currently no cure. In MERLIN, BioInVision partnered with several EU institutions developing stem cell-based therapies specifically targeting the inflammatory components of liver disease. BioInVision’s cryo-imaging technology powered by the CryoVizTM instrumentation and software was employed to provide high-resolution, microscopic colour anatomical and molecular fluorescence volumetric imaging, accurate stem cell detection, manual review and editing of cell detection, probabilistic segmentation of organs in 3D from anatomical and fluorescence volumes, global and local cell quantification, and 3D visualization of stem cell biodistribution within mouse-sized volumes. Together, MERLIN consortium partners and BioInVision set out to develop MSC therapy for liver disease.
1.3.6.2 WP6 Key Work Undertaken and Results Achieved
For MERLIN, BioInVision customized the CryoVizTM instrument to support (i) high throughput cryo-imaging of multiple-sample-arrays within the same frozen block, (ii) smart image acquisition that employs a tissue detection algorithm for skipping over unnecessary regions during imaging thereby reducing imaging time, (iii) multi-fluorophore stem cell detection that enabled tracking of multiple populations of stem cells as well as immune cells within the same specimen, and (iv) semi-automatic organ/tissue segmentation that aided in analysing local biodistribution of cells within specific organs and tissues. In software, BioInVision developed the MERLIN Biodistribution Explorer (MBE), a tool for analysing large volumes of cryo-imaging data, thereby complementing the hardware modifications to the CryoVizTM. MBE supports profile-based access privileges upon login, structured browsing of experimental data organized by work package, institution, lead investigator, and imaging study, browsing of metadata, 3D graphics and movies, and text comments relevant to an imaging study, and live data viewing in 2D/3D in an interactive fashion. MBE is augmented by an excellent set of software modules that provide high-throughput stem cell detection, manual review and editing of detection results, global and local quantification of cells, probabilistic 3D segmentation of organs and tissue, and 3D visualization of stem cell biodistribution within mouse-sized volumes. Together, the modified CryoVizTM system and software were employed in several work-packages of MERLIN, which we describe in the next paragraphs.
First, in a study with UNIPD, CryoVizTM was used to investigate the mechanism of action of MSC (WP4). Mesenchymal stem cells (MSC) represent a promising therapeutic approach in many diseases in view of their potent immunomodulatory properties, which are only partially understood. Although MSC seem to be effective in treating diverse immune-mediated disorders, a unifying mechanism of action was missing. In a previous study, the proteomic analysis of the MSC secretome identified the tissue inhibitor of metalloproteinase-1 (TIMP-1) as a potential effector molecule responsible for the anti-angiogenic properties of MSC. Using CryoVizTM cryo-imaging and cell dispersion analysis module within MBE, it was verified that subcutaneously injected MSC did not migrate away from the site of injection during the experimental time (5 days), both in immunized and in untreated mice (Figure 10). Together with the previous study, these data indicate that MSC are able to dampen inflammation through the release of soluble mediators. This study of the broad anti-inflammatory properties of MSC paves the way for developing strategies that exploit MSC-mediated inhibition of lymph node angiogenesis in the treatment of inflammation-associated pathologies.
Second, in collaboration with EMC, the immunogenicity of MSC was examined (WP3). Mesenchymal stromal cells (MSC) possess immunomodulatory properties and are low immunogenic, both crucial properties for their development into an effective cellular immunotherapy. They have shown benefit in clinical trials targeting liver diseases, however the efficacy of MSC therapy would benefit from improvement of the immunomodulatory and immunogenic properties of MSC. In this cryo-imaging study, the responsiveness of ucMSC to in vitro optimisation treatment was studied. By employing 3D visualization and quantification analysis of biodistribution from whole mouse cryo-imaging volumes, we observed that four hours after intravenous infusion in mice with CCl4-induced inflammatory liver injury, the majority of ucMSC were still trapped in the lungs. Rapid clearance of ucMSC(VitB6), ucMSC(Starv+VitB6) and ucMSC(MC) and altered bio-distribution of ucMSC(TGFβ) compared to untreated ucMSC was also observed (Figure 11). This study confirmed that majority of infused ucMSC accumulate in the lungs and that there is a rapid clearance of cells, with the cells mostly moving to the liver, an important finding that would inform future cellular immunotherapy strategies.
Third, in partnership with UoB, the efficacy and biodistribution of MSC in both acute and chronic mouse models of PSC was studied (WP1). The ability of mesenchymal stromal cells (MSC) to immunomodulate offers therapeutic potential but the inherent heterogeneity of unsorted MSC populations may explain varied/reduced function and poses regulatory challenges. Thus, in this study, the efficacy of purified CD362+ MSC in relevant pre-clinical models of PSC as a prelude to a clinical trial was examined. A time-course of bio-distribution was determined by whole mouse cryo-imaging of Q-dot labelled MSC, followed by MBE analysis. A seven-day time-course indicated MSC were cleared rapidly although there was a liver-specific increase in MSC 2-3 days after infusion (Figure 12). We concluded that CD362+ human MSC exert a potent anti-inflammatory action in murine models of PSC.
Last, in partnership with ORB, homing and biodistribution of ORB1+ MSC was studied upon injection into healthy and CCl4-induced liver injury mice using CryoVizTM cryo-imaging (WP5). First, ORB1+ MSC were isolated, expanded, and cryopreserved. Subsequently, they were injected into the two groups of mice - healthy and injured - and their homing and biodistribution was studied using MBE software. In the first experiment, mouse MSC from UBC-eGFP mice (green) were injected into FAP-DTR BAC Tg mice (low and high doses) to study homing and distribution. The aim of the experiment was to see if ablating FAP+ cells with DTX in the FAP-DTR mouse altered the engraftment distribution of the GFP+ MSC. In a second experiment, PKH26 (red)-labelled MSC were injected in uninjured and acute CCL4 injury mice to study distribution, homing and survival. It was found that cell retention and homing to affected organs was higher in the injured group as compared to the control group. (Figure 13).
Figure 10
Figure 11
Figure 12
Figure 13
1.3.6.3 WP6 Conclusion
The MERLIN project aimed to develop a therapy using purified MSC populations that alleviates the ongoing inflammation that drives fibrosis in liver. Prior to MERLIN, there was lack of evidence and concerns regarding human PA-MSC survival and persistence upon transplantation into mice. The research faced unknowns such as retention period of MSC anti-inflammatory behaviour, whether there was differentiation upon infusion, the molecular mechanisms underpinning efficacy, immuno-modulatory action in vivo, etc. Pre-clinical imaging experiments on the CryoVizTM conducted by BioInVision along with MERLIN partners through several work-packages (WP1- UoB, WP3- EMC, WP4- UNIPD, WP5-ORB) have helped provide some answers, especially pertaining to MSC homing and biodistribution upon transplantation into injured and healthy mice. This knowledge is enabling the MERLIN consortium partners to better understand the mechanism of action of the MSCs in mouse models. The study results will help optimize the MSC type to be administered, develop cell preparation methods, and optimize the dosage strategy. All the pre-clinical findings will enable a successful Phase 1/2 clinical safety trial in patients with PSC.
1.3.7 WP7 Generation of Investigational Medicinal Product Dossier and Manufacture of Clinical Grade ORB1+/- MS
1.3.7.1 WP7 Introduction
The aim for NHSBT has been to develop a GMP compliant process for the manufacture of clinical grade CD362+ve enriched Mesenchymal Stromal Cells from umbilical cord tissue and provide these cells in approved batches for a clinical trial. This work has included completion of development and engineering runs plus validation of standardized QC testing and final release criteria in place. NHSBT has also undertaken a stability study on the Drug Product. The NHSBT manufacturing facility in Birmingham, UK, has an MHRA license for this work and the results informed the Clinical Trial Application which included the development of an IMPD and Investigator Brochure. The Drug Product is produced directly from the Drug substance and cryopreserved in aliquots of 40 x10^6 or 80 x10^6 cells per bag, ready for issue and transport to the ward and then thawing ahead of administration to patients.
1.3.7.2 WP7 Key Work Undertaken and Results Achieved
From the start of the grant NHSBT worked closely with the other partners on the pre-clinical development and small-scale cell culture work as part of a technology transfer process to enable future production of MSCs under GMP by NHSBT. The NHSBT manufacturing unit along with the NHSBT Quality Assurance Dept reviewed the process design, equipment, proposed consumables and reagents to ensure that a full-scale process could be developed in compliance with MHRA regulations. Optimisation of QC assays, preparation of SOP and BMR documentation and equipment Installation and Operational Qualification validation studies were carried out.
Extensive work was undertaken to develop a consistent supply of Bone Marrow as a source of MSCs. As this proved challenging, the decision was taken to switch the source material for MSCs to umbilical cord (UC) tissue of which NHSBT has ready access to via the NHS Cord Blood Bank.
The development of a cell selection protocol to enrich for CD362+ cells from UC was undertaken in the first period although this subsequently required additional work to ensure the process met GMP standards. A significant achievement in period 2 was the development of a complete process for MSC production from UC tissue that could achieve MHRA approval. The development and validation of this process formed the bulk of the contribution from NHSBT towards the preparation of the initial IMPD and the Clinical Trial Application.
The manufacturing process required extensive development for the cell selection and cell expansion steps. This resulted in some delays and in particular, the process of selecting viable CD362+ cells from umbilical cord tissue proved difficult to develop under GMP conditions, with many of the materials not being available as CE marked for clinical use. In addition, a problem was later identified with a cell dissociation device used to assist the enzyme digestion of the UC tissue. Therefore, further development and validation work was undertaken during P3 to ensure that a new rocker device could be used in a temperature controlled incubator as a suitable alternative for tissue digestion ahead of the subsequent CD362+ cell enrichment. Thus, while important challenges were encountered, they were ultimately overcome.
Methodologies for testing of the Drug Substance and Drug Product ahead of QP release were finalised during P3. The testing included Endotoxin testing, cell counts and morphology for the Drug Substance and flow cytometry for MSC phenotype, sterility, mycoplasma testing and karyotyping for the Drug Product.
As the overall process to produce cells for the clinical trial was delayed a no-cost extension was agreed to run for a year from January 2018.
Following the problems with tissue dissociation a number of additional full-scale validation runs were undertaken using the new method in the GMP compliant facilities of NHSBT Birmingham. A stability programme was also developed for the Drug Product and for the transport of the Drug Product to the clinical units.
The problems identified with the tissue dissociation device and the need to develop another process for this stage of manufacturing meant there was a need to submit a revised IMPD to the MHRA. Once the approval for the amendment was received from the MHRA the manufacture of clinical grade MSCs for use in the clinical trial was finalised and NHSBT then started to prepare cryopreserved Drug Product batches with in-house QP release for clinical use from June 2018. The process takes many weeks to expand sufficient cells from the low numbers of CD362+ enriched cells derived from umbilical cord tissue. Each batch is produced from final cell cultures in multiple Cell Stack flasks after Passages through a number of smaller cell culture plates and flasks. The Clinical trial opened in late 2018 for patients with PSC and AIH and will therefore run beyond the end of the P4 extension period. NHSBT has produced many of the batches needed for the trial during P4 and is committed to producing the final batches required in 2019.
1.3.7.3 WP7 Conclusion
The Clinical trial opened in late 2018 for patients with PSC and AIH and so the primary endpoints of the grant have been met. A number of difficulties were encountered during the grant and lessons learned by NHSBT are primarily focussed on the difficulties in translating an R&D process into GMP compliant methodology. Many of the R&D starting procedures did not lend themselves easily to a standard GMP process that can be undertaken by a range of staff within a GMP facility. In addition, many of the essential reagents did not have CE marking and confirming these could be used under GMP was a time-consuming process. Linked to this, the later failure of a key piece of equipment to function effectively under GMP delayed the project extensively and resulted in the need to repeat many of the process development steps and validations. The final process developed is however now delivering the Drug Substance and Drug Product batches `required although it is not straightforward and will not lend itself easily to any future scale up.
The MERLIN project has been instrumental in creating an experienced team of staff within NHSBT capable of undertaking similar early phase tech transfer and development projects more readily in the future. The team has become more experienced and knowledgeable both in terms of GMP compliant manufacturing and also in the requirements of the Clinical team to gain success in an early phase clinical trial of a novel cell therapy.
1.3.8 WP8 Phase 2a clinical trial of ORB1+/- MSC in patients with PSC
1.3.8.1 WP8 Introduction
The ultimate goal of the MERLIN project is an open label clinical trial delivering CD362 selected umbilical cord derived, mesenchymal stromal cells, to patients with immune mediated liver disease. The clinical trial team remains ready to start this process, which is always challenging. Fundamentally this WP has remained live and with all the components ready to start the trial but technical challenges have delayed our progress. These technical challenges are inevitable when one proposes a novel experimental medicine study of a cell therapy, such as is planned. Participant safety is paramount and therefore the trial must be on hold until the manufacture of clinical-grade product is complete and all related regulatory approvals are in place. This appears to be close which is inevitably exciting for all.
1.3.8.2 WP8 Key Work Undertaken and Results Achieved
We generated the relevant regulatory paperwork for the Merlin trial including the study protocol, culminating in formal MHRA and Ethics submissions. D8.1 (Clinical Trials Authorisation document for the Medicines and Healthcare Products Regulatory Agency (MHRA)) was submitted in Period 2. This included an updated approval from the MHRA, in view of changes to the protocol.
The clinical trial is ongoing; the first patient was treated on 29 March 2019.
Figure 14 Press Release for the Start of Clinical Trial (UoB)
1.3.8.3 WP8 Conclusion
WE were able to overcome the many challenges of obtaining regulatory approval for the trial and successfully manufacturing MSC and opened the clinical trial in Dec 2018. We have recruited multiple patients and treatment thus far has gone well with no adverse effects. We will support the delivery of the trial with other resources and look forward to completing the trial by late 2020.
1.3.9 WP9 Data and Samples Management
1.3.9.1 WP9 Introduction
The multi-disciplinary nature of MERLIN meant that many different types of data were created and analysed throughout the project. To ensure that data were securely but openly shared, and that the movement of samples was efficiently tracked between partners, we established suitable online infrastructure, based on PT’s StudyVault platform technology.
1.3.9.2 WP9 Key Work Undertaken and Results Achieved
Data Management
During Period 1, working with the MERLIN data partners, a bespoke customisation of StudyVault was established for use in the project - StudyVault-MR. PT technologists worked with the data partners on the co-design (task 9.1) configuration (task 9.2) and installation and validation (task 9.3) of the system. On-going software support was also provided (task 9.4). The specific and unique requirements of MERLIN meant that very substantial customisation of the system was required, with some adjustment of the underlying architecture and data handling models.
StudyVault was established online and accessed via https://www.studyvault.eu/. All research data was held in an encrypted format on the cloud (there was no personal patient data created or shared on the system). Data security and confidentiality were key features. While all MERLIN data was pre-clinical research data, and so no personal or human-related data was stored, StudyVault is engineered to support compliant, secure storage of clinical data.
Co-design, installation and validation of the bespoke system was completed in the first years of the project. On-going support and any adjustments were delivered using a continuous delivery approach (so adjustments and customisations of the data platform did not hinder the performance of the live system). Significantly we added support for large-volume data formats from molecular biology platforms.
Key deliverables D9.1 Data Platform Configuration Report and D9.2 Data Platform Verification Report were completed and submitted in period 1.
During periods 2 to 4, the WP9 team continued to provide support to data partners, uploading additional data as required and making improvements to the StudyVault-MR system.
Figure 15 Example StudyVault screen
Samples Management
The management and monitoring of the movement of samples from one partner facility to another is a critical process, which has been found to frequently cause issues in life sciences projects. A key aim of WP9 was to marshal the samples management process and ensure smooth delivery of samples requirements.
Following liaison between the partners most involved in the creation and shipping of samples, an infrastructure for samples management in MERLIN was established in period 1.
During period 2, samples were transferred and managed for pre-clinical studies and sample transfers proceeded smoothly. There was a review of samples transfer logistics and some changes were introduced. The MERLIN Master Samples Register was produced. Partners opted for direct contact as the central approach for samples management, with an additional layer of oversight provided by collecting samples transfer data (with scope for partners to highlight any problems). A samples subgroup was established within the consortium to deal with any issues arising.
Samples during period 3 were dealt with and managed for pre-clinical studies without any issues arising. Direct contact between partners remained the central approach. The Master Samples Register for period 3 captured details of samples transfers, as well as protocols for sample preparation prior to transfer, storage on arrival and handling prior to use.
The Master Samples Register was maintained throughout RP4. Throughout the project, the management of samples and their transport went smoothly, and common issues of samples tracking were avoided.
1.3.9.3 WP9 Conclusion
The establishment and maintenance of the StudyVault system addressed the key challenge of multi-format, multi-owner data management in MERLIN. The system provided the facilities and supports required by the life-sciences partners, while maintaining data security and integrity.
The samples management system also worked as planned, and ensured that MERLIN avoided common research project issues and delays around ownership, location and status of samples.
1.3.10 Main S&T Results: Conclusion
As will be evident from the above, MERLIN has generated new knowledge about the efficacy, mechanisms of action, immune response, optimisation and bio-distribution of MSCs. The project has delivered important data and lays the groundwork for the future tailored production and use of MSCs with precise characteristics for particular applications.
We have also developed advanced technologies for 3D imaging and a novel GMP-compliant manufacturing process for producing MSC-based therapeutic product (ORBCEL-C™).
We are currently in the process of delivering an early phase clinical trial, exploring the safety and effect of our therapy on inflammation in patients with PSC and AIH. The MERLIN trial will continue after the formal end of the project.
We believe the project will ultimately impact on future research, manufacturing of MSCs for clinical use and cell therapies for patients with PSC and other inflammatory diseases.
Potential Impact:
In MERLIN, WP10 was dedicated to maximising the value of the project by (i) communicating the project concept, progress and findings to a range of audiences and (ii) focusing on the future exploitation and commercialisation of MERLIN project results
1.4.1 Impact and exploitation
1.4.1.1 Overview
The MERLIN project has been looking at new ways to treat liver disease with mesenchymal stromal cells (MSCs) sourced from umbilical cord, specifically focusing on PSC and AIH as model diseases. As part of our work the team has looked at the effectiveness of MSC against inflammatory liver disease in laboratory models, explored mechanisms of action of MSC and studied optimum conditions for MSC production. We have also developed advanced technologies (3D imaging and processes for GMP-compliant manufacturing of MSCs).
Our work in the MERLIN project has culminated in the design and launch of a clinical trial, looking at safety and at the effect of MSC therapy on inflammation in patients with PSC and AIH. We hope that the MERLIN clinical trial will set the stage for future MSC-based therapies for PSC, AIH and other conditions involving inflammation.
1.4.1.2 Key Results
Some of the most significant findings from our results in the MERLIN project and our key achievements are set out below.
Efficacy:
• MSC reduce markers of liver damage and inflammation in pre-clinical models of inflammatory liver disease, as well as the number of inflammatory cells in damaged areas. MSC administration ameliorates PSC pathophysiology. Both bone marrow (BM) and umbilical cord (UC) derived MSCs show positive effects.
• MSC suppress inflammatory cells (lymphocytes) isolated from the peripheral blood and explanted livers of patients with PSC.
• Freshly thawed MSC are as effective as MSC used fresh from culture, or freshly thawed and washed (in murine models of liver damage).
• The route of infusion of UC MSC has no impact on therapeutic efficacy (MSC infused subcutaneously and intravenously were equally effective).
Mechanism of Action:
• MSCs injected under the skin do not migrate to other locations. Thus, the effects of MSC may be due to cytokines released by the cells (soluble mediators).
• The beneficial effect of MSCs on PSC is due, at least in part, to the extracellular vesicles (EVs) that MSCs secrete. Results suggest MSCs act as sensor of the surrounding inflamed environment, secreting EVs that block angiogenesis, limiting inflammation. The glycoprotein TIMP-1 plays a key role in the inhibition of angiogenesis by EVs.
• MSCs are phagocytosed (engulfed) by monocytic cells that subsequently migrate to other sites via the blood stream. Monocytic cells that have phagocytosed MSCs adapt an immunoregulatory phenotype and induce regulatory T cells.
Immune Response and Optimisation:
• UC MSCs are immuno-privileged, meaning they are less susceptible to immune recognition than MSCs from bone marrow.
• The immunogenic and immunomodulatory properties of MSCs are influenced by culture conditions. We have identified treatments that may enhance some properties of MSCs e.g. pre-treatment with a cytokine cocktail or with Budesonide. Notwithstanding this, in vivo experiments show good efficacy with untreated MSCs.
• Cell density and proliferation status of MSCs have a major effect on MSC phenotype, so MSC culturing protocols merit due consideration.
Biodistribution:
• MSCs injected under the skin do not migrate to other locations.
• There is evidence from imaging analysis that dispersal of sorted MSCs may be greater from injection site than wild-type MSCs.
• Notwithstanding the remote action of MSCs, with systemic delivery of MSCs we found a greater number of cells in the liver in CCl4 injury and MDR2-KO injury mice, as compared to uninjured mice. This suggests injured livers may produce factors that attract/retain/promote survival of MSCs.
• MSCs are largely located in the lungs after administration, with an accumulation of dead MSCs in the liver 24h after infusion.
MSC Manufacture and Clinical Trial:
• Our work to date has allowed us to develop a process for MSC production from UC tissue that is consistent and aligned with regulatory requirements. NHSBT have manufactured specially selected MSCs (ORBCEL-C™, discovered by Orbsen Therapeutics) for delivery to patients in the trial.
• The MERLIN clinical trial for AIH and PSC patients opened for recruitment on 7 December 2018.
• The first patient was treated on 29 March 2019.
1.4.1.3 Exploitation and Exploitation Plans
As will be noted from the above, MERLIN has resulted in a number of key innovations. Different partners will be exploiting the results of the project in different ways in the future. Exploitation can take many forms and may include: further research building on project results; commercial exploitation (new products or services, or the furtherance of existing ones), clinical exploitation and exploitation through the development of standards, guidelines or SOPs.
Some examples of the steps being taken/planned for the further exploitation of MERLIN results are set out below.
First, it is important to note, the partners will continue to run the MERLIN clinical trial after the project ends. This will ensure the continuation of the work started in MERLIN and forms an important part of the project’s legacy.
Other ways in which MERLIN partners are exploiting and building on MERLIN include:
• UoB and ORB are involved in a new basket trial where multiple autoimmune conditions will be treated using MSCs. This will include Rheumatoid Arthritis and Crohn’s disease (aim to open this trial in 2019).
• MW-ATTC collaboration (Midlands and Wales Advanced Therapy Treatment Centre), led by MERLIN Coordinator Professor Newsome has been awarded £7.3 million funding by Innovate UK, to deliver trials to help cell or gene therapies reach the clinical market.
• MW-ATTC collaboration: ORB were awarded £1.4M to establish GMP operations and develop ORBCEL-C for the POLARISE auto-immune basket trial.
• UoB (with industry partners) has been awarded £1.1 million by UK Research & Innovation (UKRI)’s Innovate UK, to investigate patients’ experience of cell and gene therapies (PROmics study assessing the effect of novel cell therapies on patient symptoms and quality of life).
• ORB are collaborating with NHSBT to test ORBCEL-C in the Wellcome Trust REALIST clinical trial in patients with moderate to severe ARDS – with 3 patients treated in January 2019.
• EMC propose further research into long term immunomodulatory and regenerative effects, linking the MoA of MSC to morphology of MSC (modifying phenotype prior to administration to impact effect), and customising MSC therapy for particular indications, cells types and locations.
• EMC are involved in the Oxford-Aarhus-Erasmus MC consortium named MEPEP (Role of Mechynchymal Stem cells to regenerate renal graft function after ischemia reperfusion injury using normothermic ex-situ machine perfusion in pig transplantation; funded by the Lundbeck foundation, DK) where they use the gained knowledge of MERLIN to repair damaged kidneys.
• As inflammation is a well-known hallmark of cancer, UNIPD are planning to investigate the role of MSC derived EVs in mouse models of tumor progression (collaboration underway).
• UNIPD also plan to test the efficacy of MSC-EVs in the control of pathological angiogenesis characteristic for instance of eye diseases and tumours.
• SOPs and other controlled documents developed in the project will form the basis for similar developments and trials to be undertaken by NHSBT in collaboration with other academic and commercial partners. NHSBT has already started a follow-on collaboration with ORB testing ORBCEL-C in the Wellcome Trust REALIST clinical trial in patients with moderate to severe ARDS.
• BIO have further research collaborations underway and planned using their innovative 3D imaging technology (CryoVizTM) e.g. metastatic cancer imaging in projects funded by the US National Cancer Institute through the Small Business Innovative Research schemes. Potential commercial opportunities include markets in regenerative medicine, cancer immune-therapy and gene expression.
• In terms of IP protection, ORB have filed a patent application relating to CD39+ cell selection and exosome production from ORBCEL (link here). UNIPD have also filed a patent application on new anti-angiogenic extracellular vesicles derived from MSCs (link here).
• Partners have also exploited their results by delivering numerous key conference presentations and achieving several peer reviewed publications (and a PhD).
1.4.1.4 Who will benefit?
MERLIN is a truly multi-disciplinary project, encompassing the generation of new knowledge about the bio-distribution and mechanism of action of MSCs, the development of advanced technologies (3D imaging and processes for GMP-compliant manufacturing), and the delivery of an early phase clinical trial. As such, the project will have an impact on future research, manufacturing of MSCs for clinical use and cell therapies for patients with PSC and other inflammatory diseases.
Patients with PSC and AIH, who currently have no treatment options, will benefit from the project. Through delivery of the Phase 2a trial, MERLIN will provide important safety and preliminary efficacy data for a novel MSC therapy. By focusing on understanding how the therapy works to fight inflammation, we are also learning important lessons that can be used in the future to develop MSC therapies for other types of liver disease and for inflammatory conditions in general. Patients will be informed of our results through our strong links to patient groups, in particular the PSC Support group in the UK. In addition to potential benefits for individual patients and their families, improving treatments for liver disease will also benefit healthcare systems through the reduction of costly liver transplant procedures and care costs associated with chronic liver disease. From a health economics perspective, MSC therapies for immune and inflammatory diseases offer potential for significant savings in the longer term.
Researchers in liver disease, immunology and regenerative medicine will benefit from the new knowledge generated by the project and the MERLIN trial. For researchers in liver disease and in the broader field of immunology and inflammatory conditions, MERLIN is providing new data on the use of MSCs to modulate immune response and to treat inflammation. The project will also have a significant impact in the field of MSC research. We are disseminating our findings widely through publications, presentations at scientific conferences and interactions with other projects. Importantly, MERLIN is part of a larger group of sister projects with many shared partners, providing an extensive network of researchers with whom we can share ideas and results. An additional benefit derived from the project will be the advancement of 3D imaging technology, which offers researchers new, better ways to carry out bio-distribution studies.
The European cell therapy industry will also benefit from the project. We have established GMP compliant processes and quality control tests to manufacture MSCs that meet the current (and emerging) regulations for MSC therapies. In addition, our research has explored the production of optimised MSCs. The project has delivered important data and lays the groundwork for the future tailored production and use of MSCs with precise characteristics for particular applications. In order to ensure that our innovations are of use to the industry, we have focused on ensuring that our advances meet regulatory requirements.
The MERLIN consortium is committed to building on the project in future collaborations, generating long-term research value and delivering societal and economic benefits through the development of advanced MSC therapies.
1.4.2 Dissemination
A summary of the dissemination activities carried out during the MERLIN project are set out below.
1.4.2.1 Website and online
The project website http://fp7merlin.eu/ was launched in month 1 of the project and has played a central role as a public dissemination tool and means of communication. Different sections of the website are aimed at different audiences, including scientists, industry, researchers, media and the general public. The website includes the following pages:
• Home page http://fp7merlin.eu/ - a general introduction to the project.
• Project Page http://fp7merlin.eu/project/ - a more detailed look at MERLIN, with a subpage describing the research plan and a subpage for publications (both project publications and related publications of interest).
• Partners Page http://fp7merlin.eu/partners/ - details of each partner in the consortium with links to partner websites.
• Media Page http://fp7merlin.eu/media-centre/ - basic project details, press releases, contact information for the Coordinator with a subpage for promotional materials http://fp7merlin.eu/media-centre/photos-videos-and-presentations/ where project flyers, brochures, newsletters and videos can be accessed.
• Related Projects Page http://fp7merlin.eu/related-projects/ - provides links to related and complimentary research projects.
• Contact Page http://fp7merlin.eu/contact-us/ – an online form which can be used to engage with the project team.
• News Page http://fp7merlin.eu/news/ – regular updates on the project and related developments.
The website has been regularly updated over the course of the Project.
Figure 16 Screenshot of Project Homepage http://fp7merlin.eu/ Merlin Twitter Account
In addition to the website, MERLIN has an established social media presence, posting regularly on twitter (https://twitter.com/fp7merlin) and on Facebook (https://www.facebook.com/FP7MERLIN/).
1.4.2.2 Media
The project website has a dedicated media centre, which includes basic project facts, contact details for the Coordinator, the initial project press release and a factsheet from the British Liver Trust on PSC. There is also a promotional materials subpage where the project brochure, leaflet and newsletters can be accessed, together with 4 short videos featuring Merlin researchers explaining the project and the work being undertaken. The webpage makes key information about the MERLIN project easily accessible to interested journalists and to the general public
Press releases were issued at the start of the project. Press coverage included the Irish Times (“Orbsen Therapeutics to take part in €6m liver disease clinical trial’, Irish Times 22 April 2014), as well as coverage on Health Canal and My Science websites. Prof. Antonella Viola (UNIPD) was also interviewed by Italian broadcaster TG1.
A final press release has been issued, reporting that the first patient has been successfully treated in the clinical trial. The trial press release has so far been covered by the American venue, Trial Site News.
Figure 17 UoB Press Release for start of clinical trial
1.4.2.3 Awards
Our young researchers have won awards and achieved their PhDs on the basis of their excellent work in MERLIN.
1.4.2.4 Project Materials
During the course of the project we issued several project materials which were designed to engage with a broad audience.
At the start of the project we issued the project brochure and flyer setting out MERLIN’s key aims. During the life of the project we issued 2 newsletters, proving updates on progress. The MERLIN website also hosts 4 project videos that feature Merlin researchers explaining the background to their research. A public version of the report for the MERLIN Scientific Advisory Board was prepared at the end of the first and second years of the project. These reports were made available on the project website.
Partners have been provided with hard copies of project materials for distribution locally and at events and conferences (including at events hosted by the International Society for Cellular Therapy (ISCT) and International Society for Stem Cell Research).
Finally, at the end of the project we issued the MERLIN final flyer, with a high level summary of the project’s key achievements intended for a general audience. We distributed this, with cover letters, information about the project and press releases, to interested Members of the European Parliament.
All of the MERLIN project materials can be downloaded from the website: http://fp7merlin.eu/media-centre/photos-videos-and-presentations/.
Figure 18 Screenshot of some of the materials available on the project website.
1.4.2.5 Videos
The project produced a series of videos to explain our work and its value to the public.
List of Websites:
The MERLIN website is at www.fp7merlin.eu
MERLIN can be contacted via the Project Coordinator:
Prof Philip Newsome,
Centre for Liver Research
5th Floor
Institute of Biomedical Research School of Immunity and Infection
College of Medical and Dental Sciences
University of Birmingham
Edgbaston, Birmingham B15 2TT
United Kingdom
Email: p.n.newsome@bham.ac.uk
