Synergistic Effects of Combined Targeting of Beta-Catenin/CBP and Bcr-Abl in CML Blast Crisis Stem/Progenitor Cells in Vitro and in Vivo as Analyzed By Single Cell Mass Cytometry (CyTOF)

Tyrosine kinase inhibitors (TKIs) have dramatically improved the outcome of CML. However, these agents are not able to eliminate leukemia stem cells (LSC). Additionally, TKIs have limited activity in CML blast crisis (BC). Accumulating evidence demonstrates that activation of β-catenin plays a critical role in the clonal evolution from CML chronic phase to BC and development/maintenance of CML-BC LSC. We previously demonstrated that combinations of β-catenin/CBP modulator C-82 with TKIs eradicated CD34⁺ CD38^-/⁺ and quiescent TKI-resistant CML-BC stem/progenitor cells and diminished mesenchymal stem cell-mediated microenvironmental protection in vitro (Blood, 2014, 124:401). In this study, aided with single-cell mass cytometry (CyTOF) analysis and data quantification/visualization with SPADE and viSNE, we investigated the effects of combining C82 or PRI-724 (a C82 pro-drug and first-in-class β-catenin/CBP modulator) with nilotinib in various phenotypically defined CML stem/progenitor cells in vitro or in vivo in human CML-BC xenografted NOD/SCID/IL2γ^null (NSG) mice.

Cells from heavily-treated TKI-resistant CML-BC patients were stained with multiple cell surface markers including those defining typical LSC phenotype and intracellular survival signaling molecules and subjected to CyTOF analysis. β-catenin expression was found to be increased in various CML-BC stem/progenitor cell compartments and was highest in the CD34⁺ CD38⁺ CD123⁺ Tim3⁺ (Tim-3⁺ GMP) stem/progenitor subset. viSNE visualization demonstrated that overexpression of β-catenin was associated with high levels of p-CRKL, c-Myc, p-AKT, p-Tyr, p-STAT3, and p-STAT5 in CML-BC progenitor/stem cells. Cells from TKI-resistant primary CML-BC samples were then treated with C82, nilotinib, or both for 48 hours in vitro and analyzed by CyTOF. Nilotinib, as expected, had no effect on CML stem/progenitor cells in these samples. viSNE 3D visualization and quantified SPADE tree analysis revealed that C82 and more significantly the combined regimen decreased various stem/progenitor cells, especially Tim-3⁺ GMP cells, a subset representing the candidate for CML-BC LSC. Next, NSG mice engrafted with human CML-BC^T315I/E255V cells were treated daily with PRI-724 (ALZET pump, 30 mg/kg), nilotinib (oral gavage, 50 mg/kg), or combination. At the end of a 4-week treatment, bone marrow cells were analyzed by CyTOF. viSNE 4D visualization revealed that PRI-724 and the combination strikingly reduced various CML-BC stem/progenitor subsets. SPADE tree quantification analysis demonstrated that the combination, more profoundly than PRI-724 alone, reduced the leukemia burden, CD34⁺ CD38^-, CD34⁺ CD38⁺ ,CD34⁺ CD38⁺ CD123⁺, and especially Tim-3⁺ GMP stem/progenitor cells in NSG mice, which was further confirmed by conventional flow cytometry. Furthermore, PRI-724 alone and the combination with nilotinib induced CD11b expression in various CML-BC stem/progenitor subsets, suggesting CML-BC stem/progenitor cell differentiation. Notably, these treatments significantly inhibited CD44, a β-catenin downstream target gene and an important component in the leukemia/bone marrow niche interaction in CD34⁺ CD38⁺ CD123^high and CD34⁺ CD38⁺ CD123^high Tim-3^high subsets; and Tim-3, a novel marker for myeloid LSC in the CD34⁺ CD38⁺ CD123^high Tim-3^high subset; respectively. Importantly, PRI-724 (P <0.05) and the PRI-724/nilotinib combination (P <0.01) significantly prolonged the overall survival of CML-BC-carrying NSG mice, as compared to control and the nilotinib only treated groups.

Collectively, results show that inhibition of β-catenin effectively targets CML-BC stem/progenitor cells and improves survival of NSG mice engrafted with human CML-BC cells. The combination of β-catenin and tyrosine kinase inhibitors markedly enhanced the effect even in cells resistant to TKIs. Furthermore, the combination induces differentiation, and may also interrupt the interaction of CML-BC LSC with the bone marrow niche. This study suggests that combined inhibition of β-catenin and Bcr-Abl represents a novel potential strategy for targeting CML-BC stem/progenitor cells and improving patient outcomes in CML-BC that merits further investigation in preclinical and clinical settings.