Abstract
Only a minority of AML patients are cured with currently available chemotherapy regimens. Thus, new drugs with novel mechanisms of action are urgently needed. In previous proof-of-concept studies, the halogenated ATP analog 8-chloro-adenosine (8-Cl-Ado) has shown preclinical activity against a variety of solid tumors and hematologic malignancies, favorable pharmacokinetic and pharmacodynamic profiles and feasibility in a first-in-man phase I clinical trial in chronic lymphocytic leukemia. However, little is known regarding the activity of this compound in AML. In contrast to other nucleoside analogs used for treatment of AML, 8-Cl-Ado is metabolized by adenosine kinase and its triphosphate derivative is incorporated into RNA and significantly inhibited RNA synthesis in AML cells (KG-1a and MV4-11 cell lines and primary bone marrow samples) in a dose-dependent manner (300 nM to 1 uM) after 24 h exposure as compared to vehicle-treated controls (p<0.05). Similarly, 20-80% inhibition of total RNA synthesis was observed in AML patient blasts, regardless of the harbored gene mutations or cytogenetic aberrations. In contrast, 24 h exposure to 8-Cl-Ado did not significantly inhibit DNA synthesis compared to vehicle at all concentrations tested (p>0.25 at all concentrations tested). 8-Cl-Ado was metabolized into its cytotoxic triphosphate, 8-Cl-ATP, which accumulated to intracellular concentrations of more than 600 uM after 12 h exposure to 10 uM 8-Cl-Ado. Accumulation of 8-Cl-ATP was associated with >20% reduction of endogenous ATP compared to control-treated cells (p<0.05). In regard to antileukemia activity, 8-Cl-Ado inhibited growth of several AML cell lines (MOLM-13, MOLM-14, KG-1a, MV4-11, OCI-AML3) with IC50s ranging from 0.2 uM to 1.4 uM after 72 h of treatment and that of primary AML blasts including those harboring the poor-risk FLT3-ITD mutation (IC50: 800 nM for FLT3-ITD-positive blasts). In an orthotopic mouse model where FLT3-ITD-positive MOLM-14 AML cells were xenografted, animals treated with 50 mg/kg/day through an implanted osmotic pump had >70% reduction in tumor mass compared to vehicle-treated controls after 16 days of treatment (p<0.05). Importantly, 24 h pre-treatment of human LSC-enriched CD34+CD38- blasts with 5 uM 8-Cl-Ado resulted in a significant inhibition of their colony forming ability compared to vehicle-treated controls (>50% reduction of colonies after 14 days, p<0.05). To study the effect of 8-Cl-Ado on primary AML in vivo, sub-lethally irradiated Rag-2/gamma(c) double-knockout immunodeficient mice were injected retro-orbital with ten million primary human AML blast cells pre-treated with 5 uM 8-Cl-Ado or vehicle control 24 h prior to i.v. injection. Significant longer survival was observed of mice engrafted with drug-treated cells compared to those engrafted with vehicle-treated controls (p<0.0004). Given the previously reported higher expression of the antiapoptotic BCL-2 protein and the implication that this had on treatment resistance in AML, combination strategy using 8-Cl-Ado together with the BCL-2 inhibitor venetoclax was investigated and showed strong synergistic, anti-proliferative effects in both AML cell lines and primary AML blasts carrying a variety of mutations and cytogenetic aberrations. Combination index value calculation and isobologram display demonstrated strong synergy between venetoclax and 8-Cl-Ado in all 4 AML cell lines and 2 primary AML samples investigated, at doses required to inhibit 50, 75, and 90% of cell growth (ED50, ED75, ED90). We also show that, in KG-1a and MV4-11 cells, 8-Cl-Ado and venetoclax synergize in the inhibition of intracellular ATP pools. AML cell lines were treated for 6h with 10 uM 8-Cl-Ado, 250 nM venetoclax or both, before measurement of intracellular ATP. While single agent treatment decreased intracellular ATP levels in both cell lines by about 20% (8-Cl-Ado) to 40% (venetoclax), combination treatment lowered ATP levels by about 70% (KG-1a) to 90% (MV4-11). Taken together, 8-Cl-Ado is a promising agent with a unique RNA and ATP-targeting mechanism of action, excellent toxicity profile and encouraging preclinical anti-leukemia activity in AML and synergizes with the BCL-2 inhibitor venetoclax in vitro. Based on these results, we have initiated a single-agent phase I/II clinical trial in patients with relapsed/refractory AML and planned further combination with BCL-2 inhibitors.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.