Mouse models of cancer are easy to use, reproducible, and have manageable costs. Mice can also be genetically engineered to study specific oncogenic pathways. Given the many positives associated with murine models, the vast majority of pre-clinical anti-cancer therapeutic studies are performed in a mouse. Unfortunately, the vast majority of oncology drugs (~94%) that are effective in murine tumor studies will ultimately fail when tested in humans. These failures cost billions of dollars and waste physician and patient resources. Clearly, current preclinical mouse tumor models are not accurately predicting a patients? response to therapy. Thus current mouse models need to be improved to more accurately reflect human disease. Each of the currently available mouse models of human cancer has distinct strengths, but also critical flaws. One critical flaw in all available preclinical models is the lack of human tumor stroma. However, work from our group and others demonstrates that human tumor stroma is a critical component of the human tumor microenvironment which promotes human tumor growth, metastasis, suppresses anti-tumor immunity, and induces therapeutic resistance. Our long term goal is to develop a human tumor model that will, by better representing human tumors, will better cancer therapeutic response seen in human clinical trials. We hypothesize that a human tumor model with both human cancer cells and human tumor stroma will better represent primary human disease. As such, it will be a better/more stringent model for preclinical drug screening that could ultimately prevent a drug which is destined to fail from being used in clinical trials. We further hypothesize that human tumor stroma will better maintain tumor cell heterogeneity over time to more accurately reflect the natural evolution of therapeutic resistance. Finally such a model will be ideal for evaluating the impact of human tumor stroma on novel immune cell therapies. We propose to use a combination of human adult stem cells and cancer associated stem cells to create human tumor model with both human cancer cells and stromal cells. We will perform extensive histological and molecular profiling to confirm this model reflects primary tumor. Finally, we will (i) evaluate the models ability to predict patient response to disease in a retrospective study and (ii) assess the use of this model to evaluate novel immunotherapies using a novel CAR T-cell construct targeting ovarian cancer cells.
Current mouse models of cancer used for therapeutic development do not accurately predict patient response to therapy in clinical trial. We propose to develop and validate a paradigm shifting human in mouse model of cancer with a completely humanized tumor stroma. This model will, by better mimicking human disease, serve as a better tool for therapeutic development and if successful could revolutionize preclinical drug testing.