BACKGROUND: Dasatinib is a potent, multi-targeted kinase inhibitor that was recently approved for treatment of chronic myelogenous leukemia resistant to imatinib. To aid the clinical development of dasatinib in prostate cancer, we utilized preclinical models to identify potential molecular markers for patient stratification and efficacy monitoring. RESULTS: Using gene expression profiling, we first identified 174 genes whose expression was highly correlated with in vitro sensitivity of 16 cell lines and, thus, considered as candidate efficacy predictive markers. Among these are important prostatic cell lineage markers, cytokeratin 5, androgen receptor and prostate specific antigen. Our results indicate that 'basal type' cell lines with high expression of cytokeratin 5 and low expression of androgen receptor or prostate specific antigen are sensitive to dasatinib. To identify markers as surrogates for biological activity, we treated cell lines with dasatinib and identified genes whose expression was significantly modulated by the drug. Ten genes, including that encoding urokinase-type plasminogen activator (uPA), were found to not only be potential efficacy markers but also to have reduced expression upon dasatinib treatment. The down-regulation of uPA by dasatinib was drug-specific and correlated with the sensitivity of cell lines to dasatinib. Furthermore, EphA2, a target of dasatinib, was found to be a sensitivity biomarker. CONCLUSION: Using the gene expression profiling approach and preclinical models, we have identified prostatic biomarkers that are associated with sensitivity to dasatinib. This study has provided a basis for clinical evaluation of a potential dasatinib efficacy signature in prostate cancer.