Abstract
CBFA2T3::GLIS2-positive pediatric acute myeloid leukemia (AML) remains one of the worst prognostic AML subgroups. To uncover innovative targeted therapy approaches in this disease subtype we performed genome-scale CRISPR-Cas9 screening that highlighted a strong, selective dependency on JAK2 compared to other types of cancer. Using a doxycycline-inducible JAK2 knockout (KO) system, we validated JAK2 dependency in CBFA2T3::GLIS2 cell lines, observing impaired proliferation in vitro and in vivo and apoptosis induction in vitro. Both type I (ruxolitinib) and type II (CHZ868) JAK2 inhibitors showed selective in vitro activity in CBFA2T3::GLIS2-positive AML models. To identify resistance and sensitizer mechanisms to JAK2 inhibitors, we used CRISPR-Cas9 ruxolitinib anchor screening in CBFA2T3::GLIS2 AML. sgRNAs targeting negative regulators of the MAPK pathway were enriched in the ruxolitinib-treated cells. Similarly, CBFA2T3::GLIS2 AML sublines grown to resistance under chronic ruxolitinib treatment expressed pathogenic NRAS mutations. Both approaches converged on MAPK pathway activation as a resistance mechanism to ruxolitinib treatment. Combining ruxolitinib with MEK inhibitors showed a synergistic effect in cell lines and patient-derived xenograft (PDX) cells expressing the fusion and in vivo activity in a CBFA2T3::GLIS2 AML PDX, suggesting a potential approach to target this signaling circuitry in this poor outcome AML subtype.