Cancer Engineering: Strategies to Engineer Predictable Tumor Models
Dr. Aranzazu Villasante, Department of Biomedical Engineering, Columbia University, New York
Although many drugs show promise in monolayer or in animal models systems, most fail to translate in humans and this is because they lack of ability to replicate the human microenvironment in patients. In response to these limitations, I have generated a set of predictable tissue-engineered (TE) models of cancer by using different strategies. Today, I am going to focus on some of these approaches to engineer pediatric tumors in vitro. Firstly, I will show a TE model of Ewing’s sarcoma (ES) within its bone niche. This particular strategy is based on engineered human bone by introducing osteoclasts in co-culture with osteoblasts in the 3-dimensional bone niche. This model mimics bone remodeling and recapitulates some of the features observed in the osteolytic process in cancer and also, the effects of the therapeutic reagent Zoledronic acid observed in patients. The second strategy consists in designing biomaterials with the same tumor composition to mimic the biological and mechanical properties of tumors from patients. I have developed 3D porous collagen 1-hyaluronic acid scaffolds (Col1-HA scaffolds) for studies of tumor derivedexosomes, which are known to be initiators of pre-metastatic niche formation in certain sites. Interestingly, I found high levels of a critical mediator of ES growth and metastasis (EZH2) in exosomes isolated from both patients and TE model of ES. Alternatively, we cultured TE models based on Col1-HA scaffolds into a mechanical loading bioreactor for better mimicking biomechanical forces in ES. We found that biomechanical stimuli modulate osteolytic-related proteins (i.e. RUNX2) and sensitivity to anticancer drugs, such as Sorafenib. I will also explain the use of perfusion bioreactors and cell sheet engineering to develop a novel model of Neuroblastoma (NB) to study the effect of consolidative drugs, such as Isotretinoin, on tumor vasculature and stem-like cells. Here, I will show the existence of sub-populations of NB cells with different levels of stemness properties; these levels are related to the capacity of stem-like cells to transdifferentiate and also, to chemoresistance and relapse. Finally, the take-home message of my talk will be that TE models can bridge the gap between 2D in vitro cultures and in vivo animal models in a predictive, inexpensive and low timeconsuming fashion for successfully understand cancer biology and improve cancer treatments.