Patients with ALK gene rearrangements often manifest dramatic responses to crizotinib, an ALK inhibitor. Accurate identification of patients with ALK-positive non-small cell lung cancer (NSCLC) is essential for the clinical application of ALK-targeted therapy. However, assessing EML4-ALK rearrangement in NSCLC remains challenging in routine pathology practice. The aim of this study was to compare the diagnostic accuracy of FISH, immunohistochemistry (IHC), and real-time quantitative RT-PCR (QPCR) methodologies for detection of EML4-ALK rearrangement in NSCLC and to appraise immunohistochemistry as a pre-screening tool. In this study, a total of 473 paraffin-embedded NSCLC samples from surgical resections and biopsies were analyzed by IHC with ALK antibody. ALK rearrangement was further confirmed by FISH and QPCR. ALK protein expression was detected in twenty patients (20/473, 4.2%). Of the 20 ALK-positive cases by IHC, 15 cases were further confirmed as ALK rearrangement by FISH, and 5 cases were not interpretable. Also, we evaluated 13 out of the 20 IHC-positive tissues by QPCR in additional to FISH, and found that 9 cases were positive and 2 cases were equivocal, whereas 2 cases were negative although they were positive by both IHC and FISH. The ALK status was concordant in 5 out of 8 cases that were interpretable by three methods. Additionally, none of the 110 IHC-negative cases with adenocarcinoma histology showed ALK rearrangements by FISH. Histologically, almost all the ALK-rearranged cases were adenocarcinoma, except that one case was sarcomatoid carcinoma. A solid signet-ring cell pattern or mucinous cribriform pattern was presented at least focally in all ALK-positive tumors. In conclusion, our findings suggested that ALK rearrangement was associated with ALK protein expression. The conventional IHC assay is a valuable tool for the pre-screening of patients with ALK rearrangement in clinical practice and a combination of FISH and QPCR is required for further confirmation.