Oncogenic c-ros oncogene1 (ROS1) fusion kinases have been identified in a variety of human cancers and are attractive targets for cancer therapy. The MET/ALK/ROS1 inhibitor crizotinib (Xalkori, PF-02341066) has demonstrated promising clinical activity in ROS1 fusion-positive non-small cell lung cancer. However, emerging clinical evidence has shown that patients can develop resistance by acquiring secondary point mutations in ROS1 kinase. In this study we characterized the ROS1 activity of PF-06463922, a novel, orally available, CNS-penetrant, ATP-competitive small-molecule inhibitor of ALK/ROS1. In vitro, PF-06463922 exhibited subnanomolar cellular potency against oncogenic ROS1 fusions and inhibited the crizotinib-refractory ROS1 G2032R mutation and the ROS1 G2026M gatekeeper mutation. Compared with crizotinib and the second-generation ALK/ROS1 inhibitors ceritinib and alectinib, PF-06463922 showed significantly improved inhibitory activity against ROS1 kinase. A crystal structure of the PF-06463922-ROS1 kinase complex revealed favorable interactions contributing to the high-affinity binding. Taken together, our results indicate that PF-06463922 has potential for treating ROS1 fusion-positive cancers, including those requiring agents with CNS-penetrating properties, as well as for overcoming crizotinib resistance driven by ROS1 mutation.PF-06463922 | ROS1 | kinase inhibitor R eceptor tyrosine kinases (RTKs) are vital conduits of extracellular signals that direct cell growth and survival pathways. Unregulated RTK activation through chromosomal rearrangements, point mutations, and gene amplification has been shown to be responsible for the initiation and progression of many cancers. The orphan RTK c-ros oncogene1 (ROS1) normally is expressed transiently in various tissues during development with little to no expression in adult tissues (1). Elevated full-length c-ROS1 expression levels have been observed in 20-30% of patients with nonsmall cell lung cancer (NSCLC) by gene expression profiling (2-4) and in 13% of patients with lung adenocarcinoma using immunohistochemistry (IHC) (5). However, its function, both in normal physiology and disease, remains poorly defined mainly because of its still unidentified ligand. Chromosomal rearrangements resulting in oncogenic activation of ROS1 have been observed in a subset of patients with glioblastoma (6-9), NSCLC (10-14), cholangiocarcinoma (15), ovarian cancer (16), angiosarcoma (17), inflammatory myofibroblastic tumors (18), and Spitzoid melanoma (19). To date, interchromosomal translocations or intrachromosomal deletions have resulted in the production of 20 different N-terminal ROS1 fusion genes in a variety of cancers (Table S1).ROS1 is a distinct receptor with a kinase domain that is phylogenetically related to the anaplastic lymphoma kinase/lymphocyte-specific protein tyrosine kinase (ALK/LTK) and insulin receptor (INSR) RTK families (20), suggesting that tyrosine kinase inhibitors for these receptors could have cross-activity against ROS1. A recent phase I/II cl...