Superior Activity of BET Protein Bromodomain Antagonist-Based Combination with JAK Kinase, PIM Kinase or Heat Shock Protein 90 Inhibitor Against Post-MPN-MF sAML Cells

Myeloproliferative Neoplasms with myelofibrosis (MPN-MF) demonstrate constitutive activation of the JAK-STAT signaling and often progress (~20%) to AML (sAML). As a single agent, JAK1&2 inhibitor ruxolitinib confers significant clinical benefit in MPN-MF, but exhibits modest activity in sAML, which is also incurable with standard anthracycline and Ara-C-based chemotherapy. This supports the rationale to develop and test novel combination therapies for post-MPN-MF sAML. Genetic alterations documented in sAML include those in JAK2, MPL or calreticulin (CALR) gene, as well as in TP53, TET2, ASXL1, IDH1&2, SRSF2, RUNX1, MYC, PTPN11, NRAS, and SETBP1 genes. Variant allelic frequency analyses demonstrated a common co-occurrence of JAK2V617F and mutant TP53 in the dominant clones of sAML. We and others have previously reported that treatment with BET (bromodomain and extra-terminal) protein bromodomain antagonists (BA) results in growth arrest, differentiation and apoptosis of AML cells (Mol Cancer Ther 2014,13:2315). In the present studies, we demonstrate that treatment with JQ1 but not its inactive enantiomer (R-JQ1) dose-dependently (100 to 2000 nM) mediates growth inhibition and apoptosis of the cultured (HEL92.1.7, SET2 and UKE1 cells) and primary (p) post-MPN-MF sAML cells. JQ1 treatment reduced the promoter occupancy of BRD4 and RNA polymerase II on MYC, BCL2 and IL7R promoters, attenuated the mRNA and protein expressions of BCL2, BCL-xL, MYC, CDK4/6 and PIM1, and repressed the pSTAT5 and pSTAT3 levels, while concomitantly inducing the levels of HEXIM1, p21, p27 and BIM in the sAML cells. Following engraftment of NOD/scid/IL-2Rγ null) (NSG) mice with HEL92.1.7 cell xenografts, treatment with JQ1 (50 mg/kg/day, administered IP daily x 5 days per week x 3 weeks) also significantly improved the median survival of the mice (p < 0.01). Compared to treatment with each agent alone, co-treatment with JQ1 and ruxolitinib (100 to 1000 nM) or pacritinib (250 to 1000 nM), which is also a clinically active JAK2, JAK2-V617F and Fms-like tyrosine kinase 3 inhibitor that does not inhibit JAK1, was synergistically lethal against the cultured sAML cells (Combination indices of < 1.0 on the isobologram analyses). Additionally, co-treatment with JQ1 and ruxolitinib caused a marked inhibition of pJAK2, pSTAT5, pSTAT3, MYC, CDK4/6, BCL-xL and PIM1 in the sAML cells. Co-treatment with JQ1 and the pan-PIM kinase inhibitor AZD1208 (500-3000 nM), which inhibits the PIM kinase substrates BAD, p70S6 kinase and 4EBP1, also synergistically induced apoptosis of the cultured sAML cells (CI < 1.0). HEL92.1.7 and SET2 cells are not only positive for the JAK2V617F mutation, they also express mutant TP53 (M133K in HEL92.1.7 and R248W in SET2 cells). Co-treatment with JQ1 and the heat shock protein (HSP) 90 inhibitor AUY922 (2.5-20 nM), which is known to down-regulate the levels of mutant-TP53, JAK2, c-RAF, pSTAT5, pSTAT3 and pAKT, is synergistically lethal against HEL92.1.7 and SET2 cells. We have isolated JAK inhibitor-resistant HEL92.1.7 cells (> 10-fold resistant to ruxolitinib; HEL/JIR cells) under the in vitro selection pressure of a continuous exposure to JAK inhibitor (Clin Cancer Res 2011;17:7347). Compared to the parental HEL92.1.7, HEL/JIR cells are highly and collaterally sensitive to AUY922. HEL/JIR cells also remain sensitive to JQ1-induced apoptosis. Importantly, co-treatment with JQ1 and AUY922 is also synergistically lethal against HEL/JIR cells (CI < 1.0). Taken together, these findings highlight the pre-clinical activity of BA-based combination with JAK inhibitors (ruxolitinib or pacritinib), PIM kinase inhibitor (AZD1208) or HSP90 inhibitor (AUY922) against JAK kinase inhibitor-sensitive sAML, as well as of the BA-based combination with HSP90 inhibitor or pan-PIM kinase inhibitor against JAK inhibitor-resistant sAML cells. Collectively, these findings also support further in vivo testing of these BA-based combinations against sAML cells.