404 LY2228820 Dimesylate, a P38 MAPK Inhibitor, Demonstrates Anti-Neoplastic Activity in Mouse Models of Human Ovarian Cancer

The ROS1 tyrosine kinase receptor is activated in subsets of cancers including non-small-cell lung cancer (NSCLC), glioblastoma, and cholangiocarcinoma following chromosomal rearrangement. Resulting in-frame fusions of the ROS1 tyrosine kinase domain to multiple different 5′ fusion partners lead to constitutive kinase activity and have been shown to be transforming in vitro and in vivo. The ALK/cMET tyrosine kinase inhibitor crizotinib has been recently reported to inhibit ROS1 and cause tumor shrinkage in a ROS1 fusion containing NSCLC patient. However crizotinib has poor blood brain barrier penetration potentially hindering activity against metastatic brain tumors. In addition, crizotinib has reduced potency against the gatekeeper mutant form of ALK which has been reported as a mechanism of acquired resistance in ALK positive NSCLC patients. Given the structural homology between ALK and ROS1, it is feasible that a similar mechanism of resistance in ROS1-fusion positive patients will arise in the clinic. To confirm ROS1 fusions are oncogenic drivers in ROS1 fusion containing cancer cell lines, the effect of ROS1 knockdown on growth has been determined. ROS1 siRNA leads to preferential inhibition of growth of the ROS1 fusion containing cell lines U118MG (glioblastoma) and HCC78 (NSCLC) in soft agar compared to monolayer. We also identify that AZD1480, a JAK1/2 inhibitor currently in phase 1 clinical trials, inhibits ROS1 with equivalent potency to crizotinib in vitro. Both crizotinib and AZD1480 inhibit pROS1 and decrease phosphorylation levels of downstream signaling proteins including SHP and ERK in ROS1 fusion containing cell lines in vitro and in a U118MG xenograft in vivo. However, AZD1480 demonstrates more pronounced reduction in pSTAT3 levels compared to crizotinib, likely through the added inhibition of JAK1/2. Consistent with the siRNA findings, crizotinib and AZD1480 preferentially inhibit soft agar growth of the ROS1 fusion containing cell lines and lead to regression in the U118MG xenograft model. AZD1480 also penetrates the blood brain barrier in rats following oral administration as assessed using [C]AZD1480, with a brain:blood ratio of around 0.5 after 1 hour. Further, AZD1480 inhibits the L2026M gatekeeper mutation of ROS1 within two fold potency of its activity against wild type ROS1. Together, these data confirm ROS1 as a key genetic driver of tumors containing ROS1 fusions. They also suggest that AZD1480 is a novel ROS1 inhibitor with the potential to be efficacious in ROS1 containing cancer patients including those with brain metastases from NSCLC and treat or overcome gatekeeper mutation driven resistance.