Abstract P5-02-06: Predicting the response of molecular targeting agents in triple-negative breast cancer cell lines by kinase activities

Introduction Mitogen-activated protein kinase (MAPK) and phosphoinositide 3-kinase (PI3K)-AKT pathways are two major hyper-activated cascades in triple-negative breast cancer (TNBC) that critically regulate cancer progression by enhancing cell survival, proliferation, metastasis, EMT, cancer stem cell regulate, and transformation. While many therapeutic agents targeting kinases in these pathways are being developed, the development of predictor of response for such agents are critical to successfully translate them into the clinic. Genomic analysis (amplification, deletion of mutation) is one of the prediction methods. However, these technologies do not always reflect the intrinsic functionalities/activities of the kinase molecules. Therefore, we hypothesized that kinase activity predicts the response to the targeted therapy in TNBC. Materials and methods Seventeen TNBC cell lines were used in this study. To analyze cell growth inhibition, cells were incubated for 72 h with various concentrations of trametinib or wortmannin, then processed for sulforhodamine B (SRB) staining assay. To measure MEK or PI3K enzymatic activity, TNBC cell lines were lysed and immunoprecipitated with magnetic beads conjugated with MEK antibody or with PI3K p110α antibody. Kinase reaction buffer including respective substrate and ATP was added to the immunoprecipitates and incubated for 120 minutes at 37 °C. Resultant ADP was quantified by HPLC and determined MEK and PI3K activities. Protein mass of MEK, PI3K, phospho-MEK and phospho-PI3K were determined by Western Blot analysis. Total protein amount was measured by A280. Lactate dehydrogenase (LDH) activity was measured by N-assay L LDH Nittobo. Total protein and LDH were used to normalize MEK and PI3K activities for the further analysis. Results Seventeen TNBC cell lines were classified into 4 groups depending on pattern of inhibition to two inhibitors as follows; Wortmannin (PI3K inhibitor) sensitive group (W, 2/17), Trametinib (MEK inhibitor) sensitive group (T, 2/17), Both sensitive group (S, 5/17) and Resistant group (R, 8/17). We found that ratio of PI3K activity and MEK activity showed good agreement to the cell classification (PPV [Wortmannin]: 67 %, PPV [Trametinib]: 33 %, NPV: = 100 %). The other parameters; enzymatic activity of MEK or PI3K, protein mass of MEK, PI3K, phospho-MEK, or phospho-PI3K, ratios of the protein mass, and the phospho-protein did not show statistically significant agreement to the classification. Mutational status and enzymatic activities or cell classification had no correlation. Additionally, MEK activity correlated to downstream phospho-ERK expression level (R = 0.7309). Conclusion Our results show that relative activity of two relevant kinases in the signaling cascade could predict the cell lines that will not respond to molecular targeting agents against corresponding cascades. Our concept should be warranted in the clinical study with statistically sufficient number of patients. Citation Format: Sato N, Wakabayashi M, Lee J, Lim B, Ueno NT, Ishihara H. Predicting the response of molecular targeting agents in triple-negative breast cancer cell lines by kinase activities. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P5-02-06.