Closing the gap between preclinical drug discovery research and effective clinical treatments remains one of the biggest challenges of modern medical science. In the context of cancer drug discovery, this problem is not only hindering the marketing authorisation of effective therapeutics, but is also extremely costly for drug developers and delays effective relief for suffering individuals. Successful and efficient preclinical drug testing often relies on data from in vitro cell culture models that recapitulate the structure and function of tumours and other tissues in vivo. There is growing evidence that in vitro 3D cell models can more accurately predict the efficacy of drug responses compared to the traditionally utilised 2D cell culture systems. This realisation has led to continuous scientific and technological advances that have enabled more physiologically representative in vitro modelling of in vivo tissues. Ultimately, these technologies have enabled the creation of 3D cell models that are more biologically relevant, which is a critical component in successful preclinical drug testing. However, increasing the physiological relevance of biological models is only one piece of the puzzle. In order to facilitate widespread adoption of these technologies for the purposes of drug discovery, a number of challenges need to be addressed. In order to screen the large compound libraries used in drug discovery testing, the generation of 3D cell models needs to occur in a high-throughput manner. Moreover, the generated 3D cell models need to be compatible with the industry-standard downstream analysis techniques that are currently utilised in drug screening applications. This white paper will provide an overview of the utility of current 3D cell models from a drug screening perspective and explore the future of 3D cell models for cancer drug discovery.