BMS-214662 Eliminates Quiescent and Proliferating Acute Myeloid Leukemia Cells through Activation of Protein Kinase Cβ and Enhances the Efficacy of Cytosine Arabinoside

Abstract 2167

Despite major advances in our understanding of the pathophysiology of acute myeloid leukemia (AML), the prognosis remains poor in most patients. While the majority of patients (∼80%) who tolerate induction therapy enter remission, both early and late relapses are a significant problem. It is postulated that, in some cases, these relapses are due to a population of primitive AML stem cells which are not effectively targeted by conventional chemotherapy agents such as cytosine arabinoside (Ara-C) or standard chemotherapy combinations (e.g. DA, ADE or FLAG-IDA). BMS-214662 (Bristol-Myers Squibb), a cytotoxic farnesyltransferase inhibitor (FTI), preferentially kills non-dividing malignant cells (Lee et al, Proc AACR 2001;42:260S) and in a clinical trial as a single agent had anti-leukemia activity in AML (Cortes et al, J Clin Oncology 2005;23:2805). Previous studies have shown that, in chronic myeloid leukemia (CML), BMS-214662 induces mitochondrial apoptosis of primitive progenitor cells via activation of protein kinase Cbeta (PKCβ; Copland et al, Blood 2008;111:2843; Pellicano et al, Blood 2009;114:4186). We therefore assessed the efficacy of BMS-214662, alone and in combination with Ara-C, in patient-derived CD34⁺ and CD34^- AML cells in vitro using a CFSE-based flow cytometry method to track cell division and caspase-3 activity, PARP cleavage, mitochondrial membrane potential (MMP) and measurement of reactive oxygen species (ROS) to assess apoptosis. PKCβ was assessed by Western blotting ± BMS-214662, ± bryostatin-1, a PKCβ modulator. Primary AML cells were cultured for 72 hours in serum free medium supplemented with 5 growth factors (IL-3, IL-6, Flt-3 ligand, G-CSF and SCF). Conditions studied were: (1) no drug control (NDC), (2) BMS-214662 (250nM; IC₅₀ dose), (3) Ara-C 500nM, and (4) BMS-214662 250nm + Ara-C 500nM (BMS-214662+Ara-C). Total viable cells were reduced in all treatment arms compared to NDC. The combination of BMS-214662+Ara-C produced the greatest cytotoxic effect with total viable cells reduced to <20% of NDC after 72 hours (P=0.05). While Ara-C alone did not reduce the number of undivided CFSE^max AML cells compared to NDC, both BMS-214662 alone and BMS-214662+Ara-C resulted in reductions of 47 and 75%, respectively. Furthermore, BMS-214662+Ara-C significantly reduced undivided CFSE^max AML cells compared to either BMS-214662 or Ara-C alone, (P<0.04 and P<0.03, respectively), indicating at least additive activity of this combination. In primary AML cells, BMS-214662 triggered apoptosis via the intrinsic pathway with loss of MMP, generation of ROS, cleavage of PARP and caspase-3 activation. Further experiments demonstrated that PKCβ was up-regulated in a dose-dependent manner in the presence of BMS-214662. Importantly, BMS-214662 at the concentrations effective against AML CD34⁺ cells did not initiate apoptosis of normal CD34⁺ hematopoietic cells. Finally, the addition of the PKCβ modulator, bryostatin-1, abrogated the effects of BMS-214662, resulting in reduced activation of caspase-3. These results are interesting as they identify a common mechanism for induction of apoptosis in quiescent CD34⁺ cells from different myeloid leukemias (all phases of CML and different AML subtypes) which may be unrelated to the FTI activity of this compound. In conclusion, these results indicate that BMS-214662 induces apoptosis of both quiescent and proliferating AML cells via up-regulation of PKCβ and activation of the intrinsic apoptotic pathway. The combination of BMS-214662 with conventional chemotherapy may represent a novel strategy for remission induction therapy in AML, reducing the risk of relapse.