Development of anti-cancer therapeutics has been traditionally reliant on two-dimensional (2D) systems and animal models, both of which have major limitations that contribute to the poor clinical translation of preclinical findings. The goal of this thesis work was to develop three-dimensional (3D) in vitro models of bone malignancies for accurate drug testing and mechanistic studies. To this end, I investigated the use of different 3D scaffolds to recreate the distinct in vivo bone niches relevant for these bone cancers in vitro. First, I evaluated the use of electrospun poly(ε-caprolactone) scaffolds to provide 3D architectural cues for the culture of Ewing sarcoma (EWS) cells. 3D-cultured EWS cells were remarkably different from the same cells cultured in 2D, and more similar to those grown in vivo with respect to morphology, growth kinetics, and protein expression. This work underscored the importance of providing a 3D context for tumor growth in vitro. The second part of this thesis investigated the use of 3D hyaluronan (HA) hydrogels to support the culture of bone metastatic prostate cancer (PCa) cells. Due to their high fidelity to the tumor of origin, there is an emerging interest in the use of patient-derived xenograft (PDX) models to overcome the limitations of cancer cell lines. However, existing PDX culture systems are few and limited. Hence, I sought to develop an in vitro PCa PDX model by first establishing a method to enrich for PCa PDX tumor cells, then evaluated the ability of 3D hyaluronan (HA) hydrogels to maintain the viability, morphology, growth and phenotype of the encapsulated tumor cells. This work demonstrated the feasibility of using a 3D scaffold-based approach to culture PDX tumor cells in vitro. Lastly, I incorporated integrin-binding and matrix metalloproteinase-degradable peptides to HA hydrogels to support osteoblast culture with PCa PDX cells in 3D. Through this 3D co-culture system, the in vivo structural organization, phenotype, as well as biochemical crosstalk between PCa and osteoblasts in bone was recapitulated. In this work, I demonstrate for the first time, the feasibility of co-culturing PDX tumor cells with stromal cells in vitro using a tunable 3D system for controlled mechanistic investigations.