DiscussionGabriel Benton, Ph.D., Gerald DeGray, Irina Arnaoutova, Ph.D., Hynda K. Kleinman, Ph.D., Jay George, Ph.D.
Trevigen, Inc., Gaithersburg, MD
Alex Sim, AMSBIO, Abingdon, UK
Many drugs that show potential in preclinical evaluation fail during clinical trials, and the inability of current preclinical models to properly predict drug response may be attributed to oversimplified or non-existent tumor microenvironments. To provide a more physiologically predictive model for drug screening, we have developed an in vitro, 384 well, microtumor system that presents key chemical (i.e. low pH, low glucose, low oxygen), extracellular matrix (ECM), and cellular (tumor, stroma and endothelium) attributes of tumor tissues. By using ECM proteins, we can promote the physiological architecture for each of these cell types in the microtumor. To evaluate interactions between each cell type, they are fluorescently labeled with fluorophores with different excitation and emission spectra; MCF-7 and MDA-MB-231 human breast cancer cell lines express a red fluorescence protein, while human umbilical vein endothelial cells (HUVECs) and human adipose-derived mesenchymal stem cells (hMSCs) are labeled with stable lipophilic membrane dyes. Breast cancer cells and hMSCs spontaneously assemble in vitro when co-cultured under low adhesion conditions. Once formed, these structures are deposited onto preformed endothelial tubule networks comprised of HUVECs and hMSCs and are embedded within a hydrogel composed of ECM proteins. Cellular interactions and dissemination are monitored via fluorescence microscopy, and cell proliferation of the breast cancer cells is quantified using a fluorescence plate reader. The breast cancer cells exhibit cell-cell interactions with endothelial tubules and stromal cells, forming microtumors, and breast cancer cell proliferation is reduced in the microtumors compared to 2D and 3D monocultures. Furthermore, the response to anti-cancer drugs, paclitaxel and fluorouracil, in the in vitro microtumor system is similar to the response for the same drugs in xenograft models, unlike 2D or 3D monoculture. This high throughput microtumor cancer model will provide more insight into pathological mechanisms and can better predict response to anti-cancer drug treatment.