Replacement of murine transplantation assays by 3-dimensional in vitro substitutes for the assessment of therapeutic responses of cancer stem cells.

The main objective of the proposed work is establishment of validated in vitro systems that allow replacement of murine transplantation for assessment of cancer therapeutics. Many millions of mice continue to be used worldwide for assessment of the therapeutic responses of cancers; in the UK alone, nearly half a million mice were used in 2010 for non-toxicological cancer research. Several developments in cancer research are now predicted to increase further the use of murine transplantation. These include new emphasis on the heterogeneity of tumour cell responses to therapeutics and on the roles of cancer stem cells and of epithelial to mesenchymal transition (EMT) in therapeutic resistance and tumour recurrence. A new development is the increasing clinical interest in "personalized tumour grafts" for prediction of the therapeutic responses of individual patient tumours. This approach is likely to be employed with increasing frequency despite problems of initially establishing tumorgrafts from the patient material and then of the 6-7 months needed to expand the size of the mouse population bearing a tumour up to the 150 or so mice required for the therapeutic testing. In principle, however, personalized tumour therapies could have major therapeutic benefits and development of suitable surrogate in vitro systems would, in addition to replacement of mouse tumorgrafts, provide the advantages of greater rapidity, economy, and patient benefit.

Development of an appropriate in vitro environment. The trend towards increasing use of murine assays for therapeutic analyses is due in large part to the increasing importance being placed on the need for testing therapeutic responses in what is considered a physiologically appropriate microenvironment. This is interpreted as one that takes account of both the 3-dimensional nature of in vivo tissues and the important roles of tumour interactions with stromal fibroblasts and other cells. The work contained in this proposal is based on a wide body of evidence indicating that it is now possible to use in vitro technologies to replicate a physiologically appropriate 3-dimensional microenvironment that closely mimics the microenvironment human of tumours in situ and can therefore be effectively and beneficially used to replace in vivo models for early and later stages of drug testing.

Analysis of effects of cellular heterogeneity. To fulfil the key objective of reversing current trends towards increasing use of in vivo mouse models, the proposed research will focus on analysis of effects of the in vitro environment on the differing therapeutic responses occurring within sub-sets of cancer cells. Evidence indicates that heterogeneity within carcinomas is related to the existence of at least 3 cellular phenotypes and that each phenotype may respond differently to a given therapeutic agent.

The sequential stages of the study will be: (a) to confirm that patterns of cellular heterogeneity present in 3-D organotypic models correspond to those present in tumours in situ, (b) to analyse of the roles of heterogeneity in driving invasion through the 3-D matrix, and assessment of the degree to which mechanisms and signalling pathways in vitro are similar to those controlling EMT processes in tumours in situ, (c) to design, test, and validate a heterotypic 3-D organotypic system providing reproducible measurements of experimentally-induced changes in heterogeneity, (e) to provide a system that is responsive to hypoxia and cytokines, (f) to simplify the geometry of the culture system such that it is amenable to high-throughput analysis, and (g) to provide "proof of principle" that the model is responsive to therapeutic drugs, is relatively inexpensive and rapid, and is therefore commercially viable, both for new drug discovery and for dissection of the actions of existing therapeutics on sub-populations of tumour cells.

Recent research points to important roles of sub-populations of cancer stem cells (CSCs) in the growth and spread of cancers. Of particular interest, CSCs are several-fold more resistant to therapeutic killing, a property likely to be related to tumour recurrence. This raises the need to discover drugs with effective actions in killing them. This effort has now been rendered more complex by the demonstration that epithelial-mesenchymal transition (EMT) reversibly switches CSCs between a sessile epithelial phenotype and a motile EMT phenotype. In these manifestations CSCs differ in their drug resistance patterns. It is further apparent that rates of phenotypic switching, and hence therapeutic responses, are influenced both by cell intrinsic factors and by stimuli from the tumour microenvironment.

Currently, tumour transplantation to mice is considered the "gold standard" for studies of the properties and therapeutic responses of CSCs. This is based in the assumption that environmental requirements for stem cell survival are too complex to be adequately modelled in vitro. However, mouse models are far from perfect and basic CSC properties, including therapeutic resistance and entry into EMT, are retained even in standard 2 dimensional cultures on plastic. 3-dimensional co-culture of human tumour cells with fibroblasts results in interactions and cellular changes that even more closely mimic those occurring in the in vivo situation. To replace animal transplantation for stem cell assays, we therefore propose to (a) generate a heterotypic 3D culture system, containing both tumour and stromal cells, in which the cellular responses associated with EMT mimic those present in vivo, (b) develop EMT reporter systems enabling rapid analysis of cellular responses to known inducers/blockers of EMT, (c) assess the utility of this system for high throughput assays of agents suitable for the control of EMT and for killing CSCs in their various phenotypic manifestations.

The propose work aims to drive the replacement of animal models in cancer research, through the development of appropriate in vitro models for the replacement of in vivo mouse models. Many millions of mice continue to be used worldwide for assessment of therapeutic responses; in the UK alone, nearly half a million mice were used in 2010 for non-toxicological cancer research. The use of animal models is predicted to increase markedly unless suitable surrogate in vitro assays can be developed. A key argument for the current dependence on in vivo models is that it provides a physiologically appropriate microenvironment that takes account both of the 3-dimensional nature of in vivo tissues and of the important roles of cellular interactions with stromal fibroblasts and other cells. The work contained in this proposal aims at in vitro replication of a physiologically appropriate 3-D microenvironment that mimics the in vivo tumour microenvironment and thus can be used to replace in vivo models for drug testing.

European law states that: "an experiment shall not be performed if another scientifically satisfactory method of obtaining the result sought, not entailing the use of an animal, is reasonably and practicably available". Validation of the efficacy of in vitro assays can thus be expected to lead to a legal requirement for animal replacement within the EU.