Abstract 7089: Proteomic evaluation of combination data in acute myeloid leukemia to inform mechanism of action and new targets

Lori Chan,Zuleyha Ozen,Anthony Iannetta,Eric Miele,Funda Kar,Andrew Jarnuczak,Geoff Nelson,Aleksandra Markovets,Justine Roderick-Richardson,Jelena Urosevic,Lisa Drew,Jerome Mettetal

DOI: https://doi.org/10.1158/1538-7445.am2024-7089

IF: 11.2

2024-03-22

Cancer Research

Abstract:Abstract Background: Acute myeloid leukemia (AML) is a prevalent adult leukemia. In the US, around 20,380 individuals were diagnosed with AML in 2023. Despite standard therapies, relapse and resistance persist. Cyclin-dependent kinase 9 (CDK9) has been recognized an actionable therapeutic target in AML. The potent inhibitor, AZD4573, induces durable apoptosis and suppresses tumor growth. Our study reveals that combining AZD4573 with standard-of-care (SoC) treatments deepens cell death responses in multiple AML cell models. We also used proteomics tools to understand the mechanisms underlying the vulnerability and resistance in SoC and CDK9 combinations. Methods: In the drug combination study, three SoC agents (cytarabine, decitabine, and venetoclax) were evaluated in combination using a 6 × 6 dose-response matrix in four AML cell models (MOLM13, MOLM16, MV411, and OCI-AML2). The HSA synergy score and Emax drug activity were used to assess the benefits of the drug combinations. In the proteomics evaluation study, MOLM13 cells treated for 7 days under different conditions - (1) vehicle control; (2) cytarabine alone; (3) decitabine alone; (4) venetoclax alone; (5) AZD4573 alone; (6) AZD4573+cytarabine; (7) AZD4573+decitabine; (8) AZD4573+venetoclax - were collected for proteomics analysis using LC-MS/MS. Differential expression and missing pattern analyses were used to identify dysregulated proteins. Functional enrichment analysis was employed to identify dysregulated biological processes and pathways in response to different treatments. Results: Our drug combination screens demonstrated the efficacy of AZD4573 in combination with cytarabine, decitabine, and venetoclax in 3 of 4 AML cell models. In our proteomic analysis, we took two approaches. First, we compared the proteomes of AML cells treated with SoC agents, identifying shared downregulated pathways such as serine biosynthesis, notably pointing to the rate-limiting enzyme PHGDH as a potential therapeutic target in AML. Cytarabine and decitabine induced an inflammatory response, while venetoclax increased the expression of cell division regulators like Aurora kinase B (AURKB). Interestingly, targeting AURKB with AZD2811 has been shown to overcome venetoclax resistance in AML. Second, we compared proteome changes induced by various AZD4573 combinations. We identified IRF2BP2 as a shared vulnerable protein across the three AZD4573 combinations. POUF2F2 expression was significantly induced by cytarabine but suppressed when combined with AZD4573. Given that POUF2F2 activates the PDK1-Akt pathway, our findings suggest that AZD4573-induced POUF2F2 downregulation suppresses the PDK1-AKT pathway, inhibiting tumor progression. Conclusions: Our proteomics analysis identifies vulnerable and resistant pathways and proteins, shedding light on the mechanism of action and revealing potential new targets for AML. Citation Format: Lori Chan, Zuleyha Ozen, Anthony Iannetta, Eric Miele, Funda Kar, Andrew Jarnuczak, Geoff Nelson, Aleksandra Markovets, Justine Roderick-Richardson, Jelena Urosevic, Lisa Drew, Jerome Mettetal. Proteomic evaluation of combination data in acute myeloid leukemia to inform mechanism of action and new targets [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7089.

oncology