Activation of p53 By Novel MDM2 Antagonist RG7388 Overcomes AML Inherent and Acquired Resistance to Bcl-2 Inhibitor ABT-199 (GDC-0199)

Background: Acute myeloid leukemia (AML) is characterized by the clonal expansion of immature myeloid cells. AML is treated primarily with chemotherapy, but the 5-year survival rate has only marginally increased over the past few decades, highlighting the need for novel therapies to achieve higher cure rates with more acceptable toxicities.

Bcl-2 family proteins, together with TP53, are the central regulators of apoptosis. Overexpression of Bcl-2 protein is associated with leukemia progression and chemoresistance. We have observed elevated expression of Bcl-2 in AML and recently demonstrated that Bcl-2 inhibition by ABT-199 effectively induced apoptosis in AML (Pan, et.al., Cancer Discovery, 2014). However, resistance to ABT-199 was observed in cells expressing high levels of Mcl-1 or Bcl-xL. Moreover, a recent study showed heterogeneous but overlapping expression of Bcl-2, Mcl-1, and Bcl-xL proteins in 577 AML patient samples (Bogenberger, et. al., Leukemia, 2014).

Although common in solid tumors, p53 mutations are relatively rare in AML. However, p53 functions are diminished by overexpression of MDM2 protein, an E3 ubiquitin ligase of p53 and an inhibitor of p53 transactivation. We previously reported MDM2 overexpression in 53% of primary AMLs (Kojima et al., Blood, 2005). Our group also demonstrated that p53 activation by Nutlins, the prototypical MDM2 inhibitors, induced apoptosis and growth inhibition in AML.

Rationale: Since p53 activation by MDM2 inhibitors upregulates pro-apoptotic Bcl-2 proteins like NOXA, PUMA, and Bax, which counteract Mcl-1 and Bcl-xL, we hypothesized that the second-generation MDM2 inhibitor RG7388 could overcome AML resistance to Bcl-2-specific ABT-199, and that the combination could synergistically enhance apoptosis in AML.

Results: We first demonstrated that RG7388 induced apoptosis exclusively in p53 wild type (wt) cells. RG7388 was essentially ineffective in p53 mutant or null AML cell lines such as HL-60, KG1 and THP1 (48h IC₅₀s > 5 μM). Nonetheless, it showed high potency against p53 wt cell lines (48h IC₅₀s: MOLM13 = 21.7 nM, MV-4-11 = 29.2 nM). Furthermore, stable knockdown of TP53 rendered the wt cell lines completely resistant to RG7388 (IC₅₀s > 5 μM), confirming TP53-specificity. To study if RG7388 was able to overcome inherent resistance to ABT-199, we tested its efficacy on OCI-AML3 cells, which are inherently resistant to ABT-199, AraC and Idarubicin. As a single agent, RG7388 potently killed OCI-AML3 cells (48h IC₅₀ = 148 nM). Importantly, RG7388 was ~20-fold more effective in OCI-AML3 cells than its predecessors Nutlin-3a and RG7112. We also examined the time- and dose-response of RG7388 in several genetically diverse AML cell lines (p53 wt) and found that 100 nM RG7388 was able to induce apoptosis and inhibit cell growth within 12 h.

Next we studied whether RG7388 synergizes with ABT-199 to kill the refractory OCI-AML3 cells. A combination index of 0.35 (Chou-Talalay method) indicated a strong synergy between the two compounds. The combination exhibited higher activity in killing OCI-AML3 cells than either agent alone (48h IC₅₀s: ABT-199 = 1680 nM, RG7388 = 148 nM, ABT+RG = 28 nM). Similar synergy was observed in additional AML cell lines and in primary samples. Next, we generated ABT-199 resistant cells by continuous exposure of initially sensitive AML cells to escalating concentrations of ABT-199. While 1000 nM ABT-199 had no effects on the viability of these cells, additional treatment with 30 nM RG7388 effectively killed them. This finding suggested that RG7388 was able to overcome acquired resistance to ABT-199. The mechanisms underlying this resensitization and its synergism with ABT-199 are under investigation using in vitro and in vivo model systems.

Conclusions: The novel MDM2 inhibitor RG7388 induces growth arrest and apoptosis selectively in p53 wt AML cells. Importantly, the combination of RG7388 with ABT-199 synergistically induced apoptosis in AML cell lines and primary patient cells, and RG7388 was able to overcome inherent or acquired resistance to ABT-199. Since both Bcl-2 and MDM2 overexpression are associated with poor prognosis in AML, the proposed combination of the two clinical-stage compounds could have considerable clinical potential. We will report on ongoing experiments with primary AML cells in NSG mice to determine the potential of this combinatorial approach to eliminate AML stem cells.