Abstract 1659: The mechanism of action of clinically effective dual Abl/Aurora kinase inhibitors does not involve BCR-ABL inhibition

Abstract

Abstract Characterizing the mechanism of action of clinically-active multikinase inhibitors can inform disease biology and facilitate the development of more effective therapeutics. While efforts to treat chronic myeloid leukemia patients with approved BCR-ABL tyrosine kinase inhibitors have been successful, the T315I mutation at the “gatekeeper” position confers a high degree of resistance to these agents. To date, a number of multikinase inhibitors exhibiting a broad spectrum of cellular targets have demonstrated preclinical and clinical activity against this mutation. Notably, most of these agents were designed to inhibit Aurora kinases. To determine the contribution of Aurora inhibition in BCR-ABL-expressing cells, we compared the cytotoxicity of Abl/Aurora inhibitors on interleukin-3 dependent Ba/F3 cells with BCR-ABL-transformed Ba/F3 cells and observed no therapeutic index. We then exposed parental Ba/F3 cells to N-ethyl-N-nitrosourea and selected for cells with the ability to proliferate in the presence of one of these agents (XL228) and searched for mutations in the Aurora A and B genes. One recurring, heterozygous mutation in Aurora B was identified in all XL228-resistant clonal populations, Y161N. No mutations were identified in the kinase domain of Aurora A. Notably, Ba/F3 cells harboring the AuroraB/Y161N mutation were resistant to two other clinically-effective Abl/Aurora inhibitors, VX-680 and PHA-739358. This mutation confers resistance to these inhibitors at the biochemical level as assessed through phospho-histone H3. We then assessed the sensitivity of BCR-ABL-transformed Ba/F3-Y161N cells to PHA-739358. Consistent with an effect on BCR-ABL activity, these cells were substantially more sensitive to this agent than non-BCR-ABL-transformed Ba/F3-Y161N cells. However, a mutation that is highly biochemically resistant to PHA-739358 (BCR-ABL/F317L) demonstrated the same degree of sensitivity to the compound as native BCR-ABL in cell viability assays, suggesting that PHA-739358 effects cytotoxicity of BCR-ABL-transformed Ba/F3 cells primarily through inhibition of Aurora B kinase, and secondarily through a non-BCR-ABL target that is critical for viability in the presence of BCR-ABL. Additional mutagenesis screens designed to identify this target are ongoing. We conclude that Aurora B is the principal target of the clinically-active Abl/Aurora inhibitors XL228, VX-680 and PHA-739358 in BCR-ABL-expressing cells. Cells that harbor a resistance-conferring mutation in Aurora B reveal that BCR-ABL transformation creates an increased reliance upon signaling through another target of PHA-739358, distinct from BCR-ABL. This work therefore has the potential to shed light on BCR-ABL signaling pathways necessary for cell proliferation/survival, as well as for the development of adjunctive therapies for patients with BCR-ABL-expressing leukemia. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 1659.