ROS1 rearrangement characterizes a small subset (1%–2%) of non-small cell lung cancer and is associated with slight/never smoking patients and adenocarcinoma histology. Identification of ROS1 rearrangement is mandatory to permit targeted therapy with specific inhibitors, demonstrating a significantly better survival when compared with conventional chemotherapy. Detection of ROS1 rearrangement is based on in situ (immunohistochemistry, fluorescence in situ hybridization) and extractive non-in situ assays. While fluorescence in situ hybridization still represents the gold standard in clinical trials, this technique may fail to recognize rearrangements of ROS1 with some gene fusion partner. On the other hand, immunohistochemistry is the most cost-effective screening technique, but it seems to be characterized by low specificity. Extractive molecular assays are expensive and laborious methods, but they specifically recognize almost all ROS1 fusions using a limited amount of mRNA even from formalin-fixed, paraffin-embedded tumor tissues. This review is a discussion on the present and futuristic diagnostic scenario of ROS1 identification in lung cancer.