Prostate cancer represents a major clinical challenge, particularly in its advanced and metastatic forms, for which therapeutic options remain limited and clinical outcomes are poor. Despite extensive efforts in oncology drug discovery, many candidate compounds fail during clinical development, often due to insufficient efficacy and an incomplete understanding of target-specific molecular mechanisms. In this context, computational approaches are increasingly employed to support early-stage drug discovery; however, traditional structure-based methods frequently rely on static protein representations that neglect conformational flexibility. This work focuses on the fatty acid-binding protein 12 (FABP12), a protein implicated in aggressive prostate cancer progression, with the aim of improving the computational evaluation of ligand–protein interactions through a dynamics-informed workflow.