Nusser, K. | bioRxiv Science

bioRxiv2024

TKI Type Switching Overcomes ROS1 L2086F in ROS1 Fusion-Positive Cancers

PurposeDespite the robust efficacy of ROS1 tyrosine kinase inhibitors (TKIs) in ROS1-positive non-small cell lung cancer, TKI resistance continues to hamper durability of the therapeutic response. The resistance liabilities of next-generation ROS1 TKI are sparsely characterized. DesignWe compared the activity of type I TKIs (crizotinib, entrectinib, taletrectinib, lorlatinib, and repotrectinib) to the type II TKIs (cabozantinib and merestinib), and to the type I FLT3 inhibitor, gilteritinib, in CD74-ROS1 wildtype and F2004C, L2026M, G2032R, or L2086 mutant Ba/F3 cells. The findings from the Ba/F3 cell model were confirmed using NIH3T3 colony formation assays and in vivo tumor growth. CRISPR/Cas9 gene editing was used to generate isogenic wildtype and L2086F mutant TPM3-ROS1 expressing patient-derived cell lines. These lines were used to further evaluate TKI activity using cell viability and immunoblotting methods. Molecular modeling studies enabled the characterization of structural determinants of TKI sensitivity in wildtype and mutant ROS1 kinase domains. We also report clinical cases of ROS1 TKI resistance that were treated with cabozantinib. ResultsROS1 L2086F mutant kinase is resistant to type I TKI including crizotinib, entrectinib, lorlatinib, repotrectinib, taletrectinib, while the type II TKI cabozantinib and merestinib retain activity. Additionally, we found that gilteritinib, a type I FLT3 inhibitor, inhibited wildtype and L2086F mutant ROS1, however ROS1 G2032R solvent front mutation imposed resistance. The specific binding poses adopted by cabozantinib in the DFG-out kinase conformation and gilteritinib in the DFG-in kinase, provide rationale for their activity in the ROS1 mutants. Clinical cases demonstrated response to cabozantinib in tumors developing TKI resistance due to the ROS1 L2086F mutation. ConclusionCabozantinib and gilteritinib effectively inhibit ROS1 L2086F. Clinical activity of cabozantinib is confirmed in patients with TKI-resistant, ROS1 L2086F mutant NSCLC. Gilteritinib may offer an alternative with distinct off-target toxicities, however further studies are required. Since cabozantinib and gilteritinib are multi-kinase inhibitors, there is a persistent unmet need for more selective and better-tolerated TKI to overcome ROS1 L2086F kinase-intrinsic resistance. Translational relevanceROS1 L2086F is an emerging recurrent resistance mutation to type I ROS1 TKIs, including later generation TKIs. Here, we show corroborating preclinical and clinical evidence for the activity of the quinolone-based type II TKI, cabozantinib, in ROS1L2086F resistance setting. In addition, we show activity of the pyrazine carboxamide-based type I TKI, gilteritinib, in ROS1 L2086F resistance, suggesting that gilteritinib could be another option for ROS1 L2086F TKI-resistant patients. This study represents the first comprehensive report of ROS1 L2086F in the context of later-generation TKIs, including the macrocyclic inhibitors.

cancer biology↗

bioRxiv2022

Discovery of oncogenic ROS1 missense mutations with sensitivity to tyrosine kinase inhibitors

Chromosomal rearrangements of ROS1 generate ROS1 tyrosine kinase fusion proteins that are established oncogenes predicting effectiveness of tyrosine kinase inhibitors (TKI) treatment. The cancer genome reveals nonsynonymous missense mutations in ROS1, however, their oncogenic potential remains unknown. We nominated thirty-four tumor-associated missense mutations in ROS1 kinase domain for functional interrogation. Immunoblotting revealed diverse impact of the mutations on the kinase, ranging from loss of function to significant increase in catalytic activity. Notably, Asn and Gly substitutions at the Asp-2113 position in ROS1 kinase domain were TKI- sensitive hyper-activating mutations, and transformative oncogenes in independent cell models. Molecular modeling revealed drastic alterations in the activation loop of ROS1D2113N compared to wildtype kinase. Proteomics studies showed that ROS1D2113N increases phosphorylation of known effectors akin to ROS1 fusions, and upregulates pathways not previously linked to ROS1, including mTORC2, JNK1/2, AP-1, TGFB1 and CCN1/2. In vivo, ROS1D2113N drove tumor formation that was sensitive to inhibition by crizotinib and lorlatinib. Taken together, these data show that select point mutations within ROS1 RTK are oncogenic, and maybe therapeutically targetable with FDA-approved TKI.

cancer biology↗