Pancreatic ductal adenocarcinoma (PDAC) remains a major unsolved health problem. As conventional treatments have shown only a modest impact on disease course, development of new therapeutic strategies based on the molecular biology of PDAC must be a high priority. The identification of relevant predictive and prognostic biomarkers which can be used to select patient subgroups that may benefit from conventional treatments and new targeted agents will be of considerable interest. We have demonstrated the ability of the metabolizing gemcitabine protein (the human Equilibrative nucleoside transporter 1 and the deoxycytidine kinase) in predicting the benefit of adjuvant gemcitabine-based therapy in resected PDAC patients. Beside these predictive biomarkers, we have evaluated different molecular factors that may impact on the likely course of this cancer. The chemokine receptor CXCR4 that has been shown to be implicated in PDAC tumorigenicity and aggressiveness could serve as a prognostic marker for survival after a curative-intent surgery and was associated with the pattern of tumour recurrence (distant versus local relapse). Our findings were validated in an independent cohort of patients. Overall our results suggested that (i) the benefit of an adjuvant gemcitabine-based therapy can be predicted based on the tumour expression of hENT1 and dCK, (ii) CXCR4 is an independent negative prognostic factor and an independent predictor of distant relapse suggesting that anti-CXCR4 targeting therapies can be a promising treatment in combination with cytotoxic chemotherapy in the adjuvant setting. These data open new perspectives for designing trials based on a molecular driven strategy.