Pancreatic ductal adenocarcinoma (PDAC) has an extremely poor response to chemo- and (modest-dose conventionally fractionated) radio-therapy. Emerging evidence suggests that pancreatic stellate cells (PSCs) secrete deoxycytidine, which confers resistance to gemcitabine. In particular, deoxycytidine was detected by analysis of metabolites in fractionated media from different mouse PSCs, showing that it caused PDAC cells chemoresistance by reducing the capacity of deoxycytidine kinase (dCK) for gemcitabine phosphorylation. However, data on human models are missing and dCK expression was not associated with clinical efficacy of gemcitabine. We recently established co-culture models of hetero-spheroids including primary human PSCs and PDAC cells showing their importance as a platform to test the effects of cancer- and stroma-targeted drugs. Here, we discuss the limitations of previous studies and the potential use of above-mentioned models to study molecular mechanisms underlying chemo- and radio-resistance.