The present invention provides diagnostic and prognostic methods for predicting the effectiveness of treatment of a
cancer patient with an EGFR
kinase inhibitor. Methods are provided for predicting the sensitivity of tumor
cell growth to inhibition by an EGFR
kinase inhibitor, comprising assessing whether the tumor
cell has undergone an epithelial to mesenchymal transition (EMT), by determining the expression level of epithelial and / or mesenchymal biomarkers, wherein
tumor cells that have undergone an EMT are substantially less sensitive to inhibition by EGFR
kinase inhibitors. Improved methods for treating
cancer patients with EGFR kinase inhibitors that incorporate the above methodology are also provided. Additionally, methods are provided for the identification of new biomarkers that are predictive of responsiveness of tumors to EGFR kinase inhibitors. Furthermore, methods for the identification of agents that restore the sensitivity of
tumor cells that have undergone EMT to inhibition by EGFR kinase inhibitors are also provided.