Despite new therapeutics options, Prostate Cancer (PCa) remains a public health challenge because of its high incidence and mortality. Limits in PCa research come from the lack of in vitro and in vivo models that mimic the human disease. Currently, 2D in vitro tissue culture models of PCa are widely used but they present numerous limits. They do not reproduce cellular morphology, tissue architecture, inter-patients and intratumor heterogeneity. Furthermore, they lack two key components of PCa tumors, the tumoral microenvironment and the cancer stem cells. In vivo murine models of PCa cannot be representative of all the genetic alterations known in prostate tumors and they hardly reproduce the pathophysiology of human metastatic progression. Consequently, the physiology of these in vitro and in vivo models do not well represent patients tumors. 3D cell cultures overcome many of these limits by sharing morphologic characteristics with in vivo tumors as well as reproducibility of in vitro models. 3D models of PCa include spheroids derived from tumor cell lines, and organoids, derived from patient. In 3D cell cultures, cell fitness is maintained, the physiological cells-cells and cell-matrix interactions are restored and an extracellular matrix surrounds the cells. Organoids, generated from PCa primary tumors or metastases, allow studies on cancer stem cells and their microenvironment. Moreover, organoids retain genetic integrity of PCa tumors. PCa organoid model is an innovative tool that offers great perspectives of therapeutic screening. In the future, organoids generated from patients' biopsies may also lead to personalized medicine.