Background. The U.S. Food and Drug Administration‐approved method for detecting EML4‐ALK rearrangement is fluorescence in situ hybridization (FISH); however, data supporting the use of immunohistochemistry (IHC) for that purpose are accumulating. Previous studies that compared FISH and IHC considered FISH the gold standard, but none compared data with the results of next‐generation sequencing (NGS) analysis. Materials and Methods. We studied FISH and IHC (D5F3 antibody) systematically for EML4‐ALK rearrangement in 51 lung adenocarcinoma patients, followed by NGS in case of discordance. Results. Of 51 patients, 4 were positive with FISH (7.8%), and 8 were positive with IHC (15.7%). Three were positive with both. NGS confirmed that four of the five patients who were positive with IHC and negative with FISH were positive for ALK. Two were treated by crizotinib, with progression‐free survival of 18 and 6 months. Considering NGS as the most accurate test, the sensitivity and specificity were 42.9% and 97.7%, respectively, for FISH and 100% and 97.7%, respectively, for IHC. Conclusion. The FISH‐based method of detecting EML4‐ALK rearrangement in lung cancer may miss a significant number of patients who could benefit from targeted ALK therapy. Screening for EML4‐ALK rearrangement by IHC should be strongly considered, and NGS is recommended in borderline cases. Two patients who were negative with FISH and positive with IHC were treated with crizotinib and responded to therapy. The U.S. Food and Drug Administration‐approved method for detecting EML4‐ALK rearrangement in lung cancer is fluorescence in situ hybridization (FISH); however, FISH may miss a significant number of patients who could benefit from targeted ALK therapy. Screening for EML4‐ALK rearrangement by immunohistochemistry should be strongly considered, and next‐generation sequencing is recommended for borderline cases.