DNA fluorescence in situ hybridization (FISH) technology is used to study chromosomal and genomic changes in fixed cell suspensions and tissue block preparations. The technique is based on specific hybridization of small labeled DNA fragments, the probes, to complementary sequences in a target DNA molecule. Demand for FISH assays in formalin-fixed, paraffin-embedded tissues has been increasing, mainly in conditions in which diagnosis is not achieved in cell smears or tissue imprints, such as solid tumors. Moreover, the development of molecular targeted therapies in oncology has expanded the applicability of tests to predict sensitivity or resistance to these agents. The efficient use of tyrosine kinase inhibitors (TKI) of the epidermal growth factor receptor (EGFR) as therapeutical agents in advanced non-small cell lung cancer (NSCLC) depends on identification of patients likely to show clinical benefit from these specific treatments. The EGFR gene copy number determined by FISH has been demonstrated as an effective predictor of outcome from NSCLC patients to EGFR TKIs; however there are pending challenges for standardization of laboratory procedures and definition of the scoring system. This methodology article focuses on the EGFR FISH assay. It details the scoring system used in the studies conducted at the University of Colorado Cancer Center in which a significant association was found between increased EGFR copy numbers and clinical outcome to TKIs, and proposes interpretative guidelines for molecular stratification of NSCLC patients for TKI therapy.