Co-Targeting Intrinsic and Extrinsic Apoptosis to Maximize Cell Death Induction in Venetoclax-Resistant AML Cells

Bcl-2 family protein-regulated intrinsic and IAP-family protein-regulated extrinsic pathways are two major apoptotic cell death mechanisms. Components of both pathways are regulated by the tumor suppressor p53. Although combinations of Bcl-2 inhibitor venetoclax (VEN) and a hypomethylating agent induce high response rates in AML, most patients ultimately relapse. In addition, pre-clinical and clinical studies have shown that TP53-mutant AML cells are less sensitive to VEN (Carter BZ, ASH 2019 and 2020; DiNardo CD, Blood 2020). We investigated if simultaneous inhibition of Bcl-2 and IAPs and activation of p53, via MDM2 inhibition could maximize apoptosis induction in AML cells with acquired resistance to VEN-based therapy or in those carrying TP53 mutations.

We treated MV4-11 cells with acquired resistance to VEN (VEN-R) with the Bcl-2 inhibitor APG2575, IAP inhibitor APG1387, or MDM2 inhibitor APG115, and with their combinations. As expected, VEN-R cells were more resistant to APG2575 compared to control cells. APG1387 alone had limited activity in both VEN-R and in control cells and the combination of APG2575 and APG1387 enhanced cell killing (P < 0.05 compared to each single agent). APG115 was active in both VEN-R and control cells and its activity was only increased by APG2575 in the control cells, but minimally affected by either APG2575 or APG1387 in VEN-R cells. However, maximal apoptosis induction was observed in both VEN-R and control cells when all three compounds were combined (P < 0.05 compared to any of the double combinations or single agent treatments).

We next treated NSG mice harboring PDX cells derived from an AML patient who relapsed on the VEN/decitabine therapy with APG2575 (50 mg/kg, p.o., daily), APG1387 (10 mg/kg, i.v., once/wk), APG115 (50 mg/kg, p.o., daily at wk 1 and 5), or combinations. At the end of a 5-wk treatment, significant reductions of human CD45 ⁺ cells were observed in all treatment groups. At 4 wks post treatment, decreased circulating leukemia cells were found in the triple and APG115+APG2575 combination groups. APG115 (139 d, p=0.009), APG1387 (130 d, p=0.004), or APG2575 (132 d, p=0.004) significantly extended mouse survival compared to controls (116 d). Among two drug combinations, APG115 plus APG1387 did not further prolong survival, APG2575 plus APG1387 (144 d, p<0.01) was more effective, and APG2575 plus APG115 (180 d, p<0.01) most effectively extended survival compared to each drug alone. Triple combination treated mice lived longest (185 d), which was significantly longer than APG115 plus APG1387 and APG2575 plus APG1387 but did not reach statistical significance compared to APG2575 plus APG115. Data showed that triple and APG115+APG2575 combinations were most effective, followed by APG2575+APG1387, then APG115+APG1387, APG2575, or APG115, and finally APG1387.

Finally, we treated Molm13 cells lacking TP53 or carrying TP53 mutations (R248W/R213*, R248Q, R175H, R282W, Y220C) with the three agents and their combinations. All mutant cells were insensitive to single drugs. Enhanced activity was observed when any of two agents were combined and combined inhibition of Bcl-2, IAPs, and MDM2 most effectively induced cell death in TP53 knockout and all TP53 mutant cells (P < 0.05 for the triple combination compared to any of the double combinations or single agent treatments, and double combinations compared to their respective single agent treatments). Western blot analysis showed that decreased cIAP1, cIAP2, XIAP, or p21 was observed in single agent or combination-treated cells. Only in the triple combination group, cIAP1, cIAP2, and XIAP as well as MDM2 were largely diminished and p21 was marked decreased.

In conclusion, our study demonstrates that co-targeting intrinsic and extrinsic apoptosis maximizes cell death induction in AML cells with acquired resistance to VEN or with TP53 deletion/mutations by antagonizing Bcl-2, eliminating cIAPs and XIAP, as well as MDM2 and p21, a finding that needs to be validated clinically.