Validation of an in vitro humanised 3D haematopoietic system to investigate haematological malignancies.

Haematopoietic stem cells (HSCs) are specialised cells with the ability to either renew themselves and produce more stem cells or to mature - a process called differentiation which produces mature blood cells. HSCs reside in the bone marrow (BM) 'niche', a microenvironment, which supports these processes. The progression and development of myeloid leukaemias such as chronic myeloid leukaemia (CML) and acute myeloid leukaemia (AML) is encouraged by the establishment of a self-enforcing leukaemic stem cell (LSC) niche, where LSCs remodel the healthy BM niche to their advantage. Over time haematological transformation, especially of myeloid origin, often involves extramedullary haematopoiesis in the spleen and liver, where LSCs move out of the BM and reside in these organs where they proliferate and expand. To date is has proved difficult to target LSCs with current therapies and to study exactly how LSCs manage to dominate and alter the niche. Advances have been hindered by the lack of robust in vitro models that recapitulate the haematopoietic system, resulting in the majority of studies being performed in mouse models. Such models are not ideal and lack a tumour-specific and/or species-specific microenvironment. This project will create an innovative long-term fully humanised in vitro haematopoietic system to model the BM and splenic microenvironments to study haematological malignancies. Using tissue engineering we have developed an artificial haematopoietic system comprising of BM and spenic niches which includes resident cell types; mesenchymal stem cells, HSCs, LSCs and splenic fibroblasts. Our system uses advanced fibronectin-based hydrogels designed to match the mechanical properties of the BM with controlled stiffness and protease-degradability mimicking the BM structure. By utilising this model system, we will provide insight into the microenvironmental changes that LSCs confer and how these changes influence the resident cells. We aim to validate this system by directly comparing normal HSCs and LSCs behaviour and changes directly to existing in vivo xenograft data, already generated in the laboratory, using the same primary samples from CML and AML patients. This technology aims to drive the REPLACEMENT of animals used for patient-derived xenograft studies and ultimately aid knowledge and treatment of a number of haematological disorders. There is a real need within the haematology field to find an alternative to REPLACE and REDUCE reliance on mouse models. Currently there are no fully developed in vitro models of the BM in the literature. Our solution represents a robust approach to recapitulate important aspects of the BM and spleen microenvironment, whilst sustaining cell survivability, allowing us to measure differences between healthy and diseased HSCs incorporated within a microfluidic device. If adopted by other researchers and industry this system has the potential to have a substantive 3Rs impact by replacing early and substantially reducing the reliance on animal models for later pre-clinical testing of LSC-targeted therapies in leukaemia and other haematological disorders in the future.

This project is highly multidisciplinary involving expertise in HSC biology, biomaterials science and microfluidics. This project will assess the utility of a newly developed in vitro humanised 3D haematopoietic system as a relevant alternative to REPLACE mouse models for studying haematological disorders especially myeloid leukaemia. At present, no single model system exists, that accurately replicates the human haematopoietic system. The gold standard are in vivo animal models, usually involving xenotransplantation of human leukaemic cells into highly immunocompromised mice. However, these models are not ideal, with high variability observed between patient samples, low engraftment rates and often the leukaemia, which develops not accurately replicating the human disease. Even then, high numbers of LSCs are required resulting in low engraftment rates ~20% for chronic and 40-66% for acute myeloid leukaemia with <25% developing leukaemia. These assays also entail serial transplantation to assess the self-renewal of HSC/LSC, requiring multiple mice for prolonged periods of time. Another issue is the lack of an immune system and a tumour-specific and/or species-specific microenvironment in the experimental host. In fact, in myeloid leukaemia the interactions between LSCs/HSCs and the specific microenvironment may be essential for self-renewal of LSCs and thus disease evolution. Our system has several advantages; it is permissible to both normal HSC and malignant LSC proliferation and survival enabling normal and malignant haematopoiesis to be studied in parallel, and importantly patient samples, that do not engraft in vivo. In addition, it enables the early pre-leukaemic phase of the disease to be investigated and has the ability to add immune cells. Once validated this system would provide an excellent model for pre-clinical testing of novel therapies to target the LSCs, known to lead to persistent minimal residual disease (MRD) and relapse of leukaemia in patients.