Pancreatic ductal adenocarcinoma (PDAC) and gastroesophageal cancer (GEC) are among the most lethal malignancies. The absence of early symptoms means that most patients are diagnosed at late stages with advanced disease. Despite progress in treatment, the overall prognosis remains poor mainly due to their inherent and/or acquired resistance to conventional therapies. The efficacy of new treatment strategies has been hampered by the lack of efficient preclinical models recapitulating the heterogeneity and complexity of these cancers resulting in the ineffective clinical translation of novel targeted therapeutic options. Therefore, we explored the impact of traditional two-dimensional (2D) and threedimensional (3D) culture conditions on cancer stem cell (CSC) markers, epithelialmesenchymal transition, hypoxia profiles, and treatment responses in PDAC systems. We found that 3D cultures more closely mimic the tumour microenvironment and exhibit distinct cellular behaviours, suggesting that 3D organoids provide a more accurate model for studying treatment responses and therapeutic resistance. Acquired drug resistance to 5-FU (the backbone chemotherapeutic of treatment regimens for PDAC and GEC) in two patient-derived organoids (PDOs) were developed to investigate the mechanisms of chemotherapeutic resistance and identify novel targeted therapeutic strategies. Transcriptomic and proteomic analyses identified key dysregulated therapeutic vulnerabilities. Phototoxic peptide conjugates targeting the modulators of drug resistance, and pharmacological inhibitors circumvented the development of resistance. Finally, we established and characterised PDOs from treatment-naïve GEC patients and demonstrated the feasibility of using PDOs to evaluate sensitivity to chemotherapy. Here we identified actionable targets from the genomic and transcriptomic landscape of the organoids matched to their original tumour to highlight the translatability of precision medicine. Our research demonstrated that advanced 3D organoid systems represent valuable tools for modelling drug resistance and offer opportunities to discover novel therapeutic approaches to prevent the emergence of drug resistance, to improve patient outcomes through more effective and personalised treatments.