Aurora kinase A (AurkA) is a mitotic regulator that is overexpressed in multiple cancers and inhibition of AurkA induces a delay in mitotic exit.  Alisertib (MLN8237), a selective Aurora A kinase inhibitor has shown clinical activity in a phase II clinical trial in lymphomas ¹ and is being evaluated in phase III trials as a single agent and in combination. To understand the mechanism of action and resistance to MLN8237 we developed and characterised drug resistant human acute lymphoblastic leukaemia cells. Paediatric T-cell leukaemia cells, CCRF-CEM, were selected for resistance by repeat treatments of 0.4 µM and 0.8 µM MLN8237 until cells were able to proliferate in the presence of the drug; resultant sublines were designated CEM/AKA0.4 and CEM/AKA0.8, respectively. CEM/AKA0.4 and CEM/AKA0.8 sublines were 174-fold (P < 0.05) and 261-fold (P < 0.005) resistant to MLN8237 compared to the parental CCRF-CEM cells. Evaluation of cross-resistance to other agents revealed that the CEM/AKA0.4 and CEM/AKA0.8 cells were significantly resistant to: Aurora kinase B inhibitor (19- to 23-fold); multikinase inhibitor (23- to 24-fold); and a DNA damaging agent (5- to 6-fold). There was no significant cross-resistance to the tubulin-targeted agent vincristine. Since MLN8237 can induce mitotic arrest, we investigated the effect of this agent on the cell cycle profile of the resistant cells following drug treatment. At the selecting MLN8237 concentrations for CEM/AKA0.4 and CEM/AKA0.8 cells there was no major effect on the G2/M levels, compared to parental CEM cells which displayed profound cell death at these concentrations. We are currently evaluating cells for AURKA gene mutations and AurkA activity.

In summary, we have successfully developed human leukaemia cells resistant to a clinically promising Aurora Kinase A inhibitor. The resistant cells will provide a valuable resource to understand the mechanism of action and resistance to this promising agent.