Lung cancer still has one of the highest morbidity and mortality rates among all types of cancer. Its incidence continues to increase, especially in developing countries. Although the medical field has witnessed the development of targeted therapies, new treatment options need to be developed urgently. For the discovery of new drugs, human cancer models are required to study drug efficiency in a relevant setting. Here, we report the generation of a non-small cell lung cancer model with a perfusion system. The bioprinted model was produced by digital light processing (DLP). This technique has the advantage of including simulated human blood vessels, and its simple assembly and maintenance allow for easy testing of drug candidates. In a proof-of-concept study, we applied gemcitabine and determined the IC50 values in the 3D models and 2D monolayer cultures and compared the response of the model under static and dynamic cultivation by perfusion. As the drug must penetrate the hydrogel to reach the cells, the IC50 value was three orders of magnitude higher for bioprinted constructs than for 2D cell cultures. Compared to static cultivation, the viability of cells in the bioprinted 3D model was significantly increased by approximately 60% in the perfusion system. Dynamic cultivation also enhanced the cytotoxicity of the tested drug, and the drug-mediated apoptosis was increased with a fourfold higher fraction of cells with a signal for the apoptosis marker caspase-3 and a sixfold higher fraction of cells positive for PARP-1. Altogether, this easily reproducible cancer model can be used for initial testing of the cytotoxicity of new anticancer substances. For subsequent in-depth characterization of candidate drugs, further improvements will be necessary, such as the generation of a multi-cell type lung cancer model and the lining of vascular structures with endothelial cells.