Co-Targeting BCL-2 and MCL1 (via CDK9) in Pre-Clinical Models of High-Risk Acute Lymphoblastic Leukemia

Background: Acute leukemias are still a leading cause of death in children, adolescents and young adults (CAYA) due to treatment toxicity and chemoresistance. Relapsed disease has poor prognosis, especially after immunotherapeutic approaches have failed, including blinatumomab (bispecific T cell engager BITE) and chimeric antigen receptor T-cell (CAR-T) therapy. Mechanisms of relapse to immunotherapies include antigen loss/down-regulation (e.g. CD19 negative relapse) or lineage switch, whereby salvage options are limited. We have previously shown that co-targeting the pro-survival proteins BCL-2 (using the selective BCL-2 inhibitor, venetoclax) and MCL-1 (using the selective MCL-1 inhibitor, S63845) induces synergistic cytotoxicity in pre-clinical models of acute myeloid leukemia (AML)^{Moujalled et al 2019} and acute lymphoblastic leukemia (B-ALL)^{Moujalled, Hanna et al 2020} and this combination can overcome single agent resistance. Despite this, the translation of MCL-1 inhibitors to the clinic has been hampered by unacceptable toxicity. An alternative strategy to improve the therapeutic window of targeting MCL-1 is through inhibition of cyclin dependent kinase 9 (CDK9), a critical transcriptional regulator of MCL-1 expression. The CDK9 inhibitor alvocidib, has previously been demonstrated to be tolerable in children (COG Phase 1 ADVL0017 study).

Aims: 1: Assess the efficacy (cytotoxicity and Bliss synergy) of combination alvocidib and venetoclax in pre-clinical models of ALL including kinase activated, KMTA2R and CD19 negative sub-types. 2: Assess the tolerability of combination alvocidib and venetoclax in mice in vivo and healthy donor blood (stem cell and immune (innate and cellular) compartments) ex vivo. 3: Optimise BCL-2 family profiling using mass cytometry to identify patterns of pro-survival and pro-apoptotic protein expression that predict treatment response and resistance.

Methods: Venetoclax, alvocidib, S63845 (MCL1i), dexamethasone and tyrosine kinase inhibitors (TKIs) were from Selleckchem. CRISPR-CAS9 gene editing was performed to generate CD19^-/- ALL cell lines. In vitro cell viability (sytox blue exclusion) assays at 24hr (combination) and 72hr (monotherapy) were evaluated by flow cytometry using a 7-log dose concentration range and drug synergy determined using a checkerboard approach to evaluate combinations as previously described ^{(Bliss, Ann Appl Biol 1939}). Tolerability studies of venetoclax and alvocidib were performed in NSG; NOD.Cg-Prkdc^scid Il2rg^tm1Wjl/SzJ mice.

Results: Combining venetoclax and alvocidib has remarkable synergy in B-ALL cell lines, including Ph+, KMT2AR and CD19^-/- subtypes. Interestingly, CDK9i showed greater single agent cytotoxicity than MCL-1i, median LC₅₀ 400nM at 72h. To determine the tolerability of venetoclax combined with alvocidib in vivo, NSG mice were treated with venetoclax (oral gavage 50mg/kg) and alvocidib (intraperitoneal 3mg/kg) daily, Monday to Friday for 4 weeks; mice were bled weekly over 28 days to determine blood count parameters. The combination was well tolerated with comparable blood cell parameters compared to the vehicle control cohorts. In vivo experiments and mechanistic studies are ongoing including patient derived xenografts (PDXs) of hypodiploid and multiply relapsed ALL following immunotherapy.

Conclusions: Simultaneous inhibition of BCL-2 and CDK9 represents an effective approach for targeting Ph+, KMT2AR and CD19^-/- B-ALL without need for additional DNA-damaging chemotherapy or kinase inhibition. Taken together, this provides strong rationale for the clinical translation of venetoclax combined with alvocidib in patients with poor prognosis ALL, thereby offering a promising novel combination treatment for CAYA.