A compound combination screening approach with potential to identify new treatment options for paediatric acute myeloid leukaemia

Paediatric acute myeloid leukaemia (AML) is a heterogeneous disease characterised by genetics and morphology. The introduction of intensive chemotherapy treatments together with patient stratification and supportive therapy has resulted in a moderate improvement in patient prognosis. However, overall survival rates remain unacceptably poor, with only 65% of patients surviving longer than 5 years. Recently age-specific differences in AML have been identified, highlighting the need for tailored treatments for paediatric patients. Combination therapies have the potential to improve patient prognosis, while minimising harmful side-effects. In the laboratory setting, identifying key combinations from large drug libraries can be resource-intensive, prohibiting discovery and translation into the clinic. To minimise redundancy and maximise discovery, we undertook a multiplex screen of 80 apoptotic-inducing agents in paediatric AML pre-clinical models. The screen was designed using an all-pairs testing algorithm, which ensured that all pairs of compounds could be tested, while minimising the number of wells used. We identified a combination of ABT-737, a Bcl-2 family inhibitor and Purvalanol A, a CDK inhibitor, as a potential targeted therapy for AML patients with an MLL rearrangement and an FLT3-ITD. Our approach has the potential to reduce resource-intensity and time associated with the identification of novel combination therapies.