Epidermal growth factor receptor (EGFR) inhibitors have been proven to improve survival in advanced non-small cell lung cancer patients, even after failure of previous chemotherapy. However, how to identify the patients, who will benefit from this treatment, is still not known. Clinical and demographic factors, i.e., females, never-smokers, patients with Asian ethnicity, or histology of adenocarcinoma, seem all to be favorable factors for clinical outcome, but not sufficient for patient selection. Increased EGFR gene copy number detected by fluorescence in situ hybridization has consistently been shown in several retrospective studies to be a good predictive "marker," especially for EGFR tyrosine kinase inhibitors. Nevertheless, most of the data obtained so far are retrospective, and prospective validation is ongoing. The current review summarizes the clinical data based on the first generation EGFR inhibitors and discusses future strategies for exploring the role of EGFR fluorescence in situ hybridization as a selection marker for EGFR inhibitor therapy in non-small cell lung cancer.