The in vivo tumor microenvironment is composed of both various cell types and the extracellular matrix (ECM), which is known to provide the mechanical support and chemical cues to regulate tumor cell growth, motility, and their responses to chemotherapeutics. Biomaterials have been widely used to recapitulate the ECM for studying of cell-matrix interactions in different types of cancers, and different 2D or 3D high-throughput biomaterial platforms have been developed as a tool to predict tumor cells' response to chemotherapeutics. Although the 3D platforms are considered to be ideal for mimicking the complex 3D tissue architecture, they are usually difficult to fabricate and scale up for high-throughput drug testing as opposed to 2D models. We have taken a holistic approach and evaluated drug response across various drugs with different cell lines on plastic surfaces, on 2D gel surfaces and within 3D gels (either as single cells or spheroids) with different stiffnesses to identify the appropriate model for each type of drug mechanism.
We used a high-throughput PEG-PC hydrogel biomaterial platform as a 2D biomaterial drug screening system. Briefly, we co-polymerized poly(ethylene glycol) dimethacrylate (PEG) and the zwitterion 2-Methacryloyloxyethyl phosphorylcholine (PC) to form an optically clear hydrogel via radical polymerization in a silane-treated glass 96-well plate, and the hydrogel modulus can be controlled by varying the PEG crosslinker content. For the 3D biomaterial, we used 3D poly(ethylene glycol)-Maleimide (PEG-Mal) gels. PEG-Mal prepolymer solution, which contains either single cells or spheroids, was crosslinked by PEG-dithiol, which was dissolved in triethanolamine, and the PEG-Mal concentration was varied to control the modulus. For spheroid formation, single cells were encapsulated and cultured in poly(N-isopropylacrylamide) (pNIPAAm) hydrogels for 14 days to form multicellular spheroid structure. We have used these systems to test breast carcinoma cells SkBr3 and MDA-MB-231's response to a common chemotherapeutic drug (doxorubicin) and three different targeted drugs (sorafenib, lapatinib, and temsirolimus), which inhibits receptor tyrosine kinases and their downstream signaling pathways, and we use a proliferation assay to quantify the cells' inhibitory concentration at 50% or IC50.
Thus far, for most of tested drugs, we have observed that both cell lines' drug responses are more sensitive to the change in stiffness on 2D PEG-PC gels, with higher IC50s on stiffer gels, compared to 3D PEG-Mal gel. For both cell lines, their IC50s are more sensitive to the change in testing platforms when treated with sorafenib and lapatinib as opposed to treating with temsirolimus and doxorubicin, and SkBr3 cells' drug response is more sensitive than MDA-MB-231 cells. We also observed that encapsulating SkBr3 spheroids resulted in a reduction in phosphorylation level of HER3 while there was not a significant change in HER2 phosphorylation. We are currently investigating the mechanisms behind these differences across these platforms. We propose that this approach will identify appropriate models for different type of drugs, therefore reducing the unnecessary use of complex models, saving cost, and facilitating the drug development.