Venetoclax Synergizes with the RNA-Directed Nucleoside Analog 8-Chloro-Adenosine in Acute Myeloid Leukemia in Vitro and In Vivo

Relapse of acute myeloid leukemia (AML) is attributed to the persistence of quiescent leukemia stem cells (LSCs). Bcl‐2 inhibition has been shown to target primitive leukemia progenitors. Venetoclax (VEN) is a FDA-approved Bcl‐2-selective inhibitor for the treatment of AML. Although the activity of single agent VEN in AML patients (pts.) is modest, clinical efficacy in newly diagnosed, older pts. unfit for intense chemotherapy has been shown when VEN is combined with the hypomethylating agents (HMAs) azacytidine and decitabine or with low-dose nucleoside analog cytarabine. We have recently shown that VEN in combination with HMAs augments oxidative stress in AML cells and provided a molecular mechanism for the VEN-HMA-regulated NF-E2-related factor 2 (Nrf2) antioxidant pathway that could explain the results observed in early clinical studies in AML. Although about 70% of pts. initially respond to these VEN treatment regimens, about 30% of pts. do not and diminished efficacy of VEN combination treatments have been observed in pts. harboring poor-prognosis markers such as FLT3-ITD. In addition, future relapse of a percentage of pts. treated with VEN combinations is expected. Thus, novel treatment options for are urgently needed.

We previously reported that the ribose containing, RNA-directed nucleoside analog 8-chloro-adenosine (8-Cl-Ado) demonstrates cytotoxic activity against AML cells and LSCs in vitro and in vivo, without significantly affecting normal hematopoietic stem cells. Importantly, our initial, unpublished results from a phase I/II clinical trial with single agent 8-Cl-Ado in pts. with refractory/relapsed AML demonstrate encouraging clinical benefits. Moreover, we have reported that FLT3-ITD AML is particularly sensitive to 8-Cl-Ado, thus suggesting 8-Cl-Ado plus VEN as a potential novel therapeutic regimen for treatment of AML.

We here report that the VEN plus 8-Cl-Ado combination inhibited in vitro growth and induced apoptosis in AML primary cells, LSCs and cell lines significantly more compared to treatment with the individual agents. For in vitro cell growth studies, combination indices of <1 for all experimental and calculated drug concentrations demonstrated strong synergy between the two drugs in 2 human AML cell lines (MV4-11 and KG-1a) and in AML cells isolated from 2 pts. Moreover, immune compromised NSG mice engrafted with FLT3‐ITD MV4-11 cells survived significantly longer when treated with VEN (20 mg·kg^‒1·day^‒1, daily oral) plus 8-Cl-Ado (50 mg·kg^‒1·day^‒1; osmotic pump), as compared to single agent or vehicle-treated mice (p<0.006, VEN+8-Cl-Ado vs. 8-Cl-Ado; p<0.001 VEN+8-Cl-Ado vs. VEN).

LSCs depend on amino acid metabolism-driven and/or fatty acid oxidation (FAO)-driven oxidative phosphorylation (OXPHOS) for energy production. VEN is known to target LSCs through inhibition of OXPHOS by targeting amino acid uptake/metabolism. We report here that 8-Cl-Ado inhibited the FAO pathway and down-regulated the oxygen consumption rate (OCR), a marker for OXPHOS, in LSCs. However, whereas 500 nM of 8-Cl-Ado was sufficient to induce MV4-11 growth inhibition, 1 microM of 8-Cl-Ado was needed for maximum inhibitory effect on FAO. We also report that 8-Cl-Ado increased expression of the anti-apoptotic protein p53. It was previously reported that p53 induces FAO in LSCs. Knockdown of p53 by siRNA augmented the inhibitory effect of 8-Cl-Ado on FAO and OCR. Importantly, addition of VEN could completely overcome the p53-induced activation of FAO and OCR. Mechanistically, we show that 8-Cl-Ado inhibited ribosomal RNA (rRNA) synthesis, a prerequisite for cellular proliferation, through down-regulation of the transcription initiation factor 1 (TIF-IA) protein. Since TIF-1A negatively regulates p53 expression, the inhibition of TIF-1A by 8-Cl-Ado resulted in up-regulation of p53 and subsequent p53-induced upregulation of FAO and OCR, thus diminishing the suppressive effects of 8-Cl-Ado on FAO and OCR. We further show that the VEN plus 8-Cl-Ado combination strongly induced p53 signaling, as shown by activation and inhibition of downstream p21 and PCNA proteins, respectively. This combination also augmented DNA fragmentation and apoptosis in LSCs.

Thus, our data suggest that the synergy seen in AML with the VEN plus 8-Cl-Ado combination can be explained at least in part due to augmented inhibition of FAO and OXPHOS and represents a promising novel treatment for AML.