Combination of Sorafenib and Arsenic Trioxide Exerts Synergistic Anti-Leukemia Effects in Sorafenib-Resistant FLT3-ITD-Harboring AML Cells

Sorafenib, originally developed as a small molecular Raf inhibitor, has been reported by us to exert directly effect by targeting internal tandem duplication (ITD) mutations of Fms-like tyrosine kinase 3 (FLT3) gene and resulted in an impressive reduction of leukemia blasts in the peripheral blood and bone marrow in AML patients [Zhang W., et al: (2008) J Natl Cancer Inst. 100:184–98]. However, complete remission was not achieved by single agent therapy. Long term treatment with sorafenib seems inducible of the chemoresistance in AML cells lines and animal models although the MAK-ERK activity still could be profoundly inhibited by sorafenib. On the other hand, arsenic trioxide (ATO) has shown great promise in the treatment of patients with acute promyelocytic leukemia (APL). However, ATO efficacy as a single agent in clinical trials to treat non-APL AML was limited and associated with activation of MEK-ERK signaling [Bonati A., et al. (2006) Curr Pharm Biotechnol. 7:397–405]. Therefore, we hypothesized that a combination therapy with sorafenib and ATO could abrogate the ATO-induced upregulation of pro-survival signaling of MEK-ERK and lead to improvement efficacy against hematological malignancies. We first established sorafenib-resistant leukemia cells by long-term culture of murine Baf3-FLT3-ITD cells in vitro with low doses of sorafenib; or by isolation of resistant cells from SCID mice which survived following prolonged treatment with low doses of sorafenib (named Baf3-ITD-Re and Baf3-ITD-Rm cells, respectively). Anti-leukemic activity of sorafenib and/or ATO was investigated in the resistant cells and parental cells (Baf3-ITD) by determination of annexin V positivity and ERK phosphorylation after treatment with sorafenib and/or ATO. In vivo effects were investigated in a leukemic murine model by injecting Baf3-ITD-Rm into SCID mice via tail vein, and tumor progression was determined by monitoring the body bioluminescence of xenograft-bearing mice and by comparing spleen and liver size. The infiltration with leukemia cells was analyzed by histology.

Our results showed that sorafenib-induced resistant cells Baf3-ITD-Re and Baf3-ITD-Rm have 1000-fold resistance to sorafenib-mediated cell apoptosis compared with parent cells Baf3-ITD. ATO treatment upregulated the level of phosphorylated ERK in these sorafenib-resistant cells. Combined sorafenib and ATO showed synergistic effect in the resistant cells, and was even more effective in stroma cell co-culture systems (CIs were 0.81 ± 0.12 vs. 0.26 ± 0.12). In a mouse leukemia model of sorafenib-resistance, combined administration with sorafenib and ATO for two weeks synergized anti-leukemia efficacy, significantly decreased the leukemia burden and effectively decreased spleen and liver size. Furthermore, histological analyses showed profoundly reduced leukemia cell infiltration in spleen, liver and bone marrow in the combination therapy. In summary, the combination of sorafenib with ATO shows a synergistic effects in a sorafenib resistant leukemia model, which may suggest a potential therapeutic strategy in sorafenib-resistant leukemia patients.