Abstract The BCL2 protein family controls apoptosis by shifting the dynamic equilibrium state of complex formation. BH3 mimetics disrupt this balance by disturbing the interaction of a specific anti-apoptotic protein, leading to the initiation of apoptosis. However, relying solely on molecular data for the specific anti-apoptotic protein can be challenging due to the functional redundancy within the BCL2 protein family and their complex interaction network. Therefore, comprehensive profiling information considering protein complexes is essential to determine apoptotic dependencies, particularly in a clinical context. In this study, we adapted a Single-molecule Protein Interaction Detection (SPID) platform that can assess levels of protein complexes in a quantitative manner. Out of the 47 metrics included in our Apoptosis Signaling 47 panel, we utilized 22 assays to measure apoptosis dependencies in acute myeloid leukemia (AML) cells obtained from 32 patients. By combining these metrics, we developed a predictive model for the ex vivo response to ABT-199 treatment, focusing on three key biomarkers: BCL2-BAX complex, BCL2-BIM binding potential, and BCLxl-BAK level. From a receiver operating curve (ROC) analysis, the model exhibited high accuracy with a value of 0.94 area under the curve (AUC). We further validated our model by comparing the predicted response to the therapeutic outcomes in 10 AML patients according to the 2017 ELN guidelines, and achieved 100% sensitivity (4 out of 4 identified correctly) and 83.3% specificity (5 out of 6 identified correctly). Overall, our platform provides an effective method to explore apoptosis dependencies in clinical samples based on the information regarding protein complexes within the BCL2 protein family. Three key molecular biomarkers can be translated into a gauge for predicting therapeutic outcomes of BH3 mimetics, suggesting an improvement of precision medicine in AML treatment. Citation Format: Hongwon Lee, Changju Chun, Byungsan Choi, Saem Hong, Yunseo Lee, Youngil Koh, Tae-Young Yoon. Deciphering multiple protein complexes in BCL2 family to measure apoptotic dependencies in acute myeloid leukemia [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 2532.