Drug discovery in oncology is characterized by high attrition rates in clinical trials. The main reasons are lack of efficacy or unacceptable toxicity. Thus, more predictive preclinical models for drug discovery and development are urgently needed. In the field of immune-oncology, preclinical models are particularly demanding since they have to reflect the complex interplay between tumor cells and immune cells in the human body. The recent years were characterized by vast advancement in 3D in vitro models and organoid techniques. These models mimic the tumor microenvironment more realistically than traditional 2D cell cultures. They can include the main components of the tumor microenvironment, like (primary) tumor cells, immune cells and other stromal cells and allow growth and interaction of the diverse cell types in 3D. Thus, these 3D models provide a physiologically relevant platform for studying the mode of action of (immuno-) oncologic therapeutic drugs. SMAC (second mitochondrial-derived activator of caspases) mimetics, like many other anticancer drugs, can exert their effects on different cell populations within the tumor microenvironment, which can be analyzed by using 3D co-cultures as preclinical models. Fluorescence microscopy is a powerful tool to study biological processes. In live cell imaging, entire cells or subcellular structures can be monitored over time with the help of fluorescent labels. In this chapter, we describe a human 3D co-culture infiltration assay combining tumor cells embedded in a hydrogel and immune cells added on top of the hydrogel. This 3D co-culture is stable for more than 1 week and gives insights into cellular drug responses over time. We used live cell imaging/fluorescence microscopy as the main readout to quantify immune cell infiltration in 3D in response to SMAC mimetic treatment.