Abstract 3557: System-based BCL2 family protein signatures as predictive biomarkers in triple-negative breast cancer

Abstract

Abstract Breast cancer is a heterogeneous disease. More recently, the molecular basis for this heterogeneity has started to be analysed at the genomic, transcriptomic and proteomic level, allowing for the development of more personalized therapies. Cancer cells showing resistance to therapy frequently have an intrinsic deficiency in their ability to initiate or execute apoptosis [1]. The BCL2 (B-cell lymphoma 2) family of proteins controls the process of Mitochondrial Outer Membrane Permeabilization (MOMP), which is required for the activation of the mitochondrial apoptosis pathway [2]. We developed a deterministic systems model of BCL2 protein interactions, DR_MOMP that calculates the sensitivity of cells to undergo mitochondrial apoptosis [3]. Here we applied DR_MOMP in the context of triple negative breast cancer to determine whether this systems model can be used as a prognostic biomarker or patient stratification tool. To analyze the contribution of BCL2 proteins in modulating apoptosis in breast cancer we validated DR_MOMP in a panel of Triple Negative Breast Cancer (TNBC) cell lines. Using quantitative Western blotting we determined the absolute protein levels of pro-apoptotic BAX and BAK and anti-apoptotic BCL2, BCL(X)L and MCL1. We found a significant correlation between cell survival and BCL(X)L levels (ρ = 0.76) after Cisplatin/Paclitaxel treatment, but protein levels of BCL2, MCL-1, BAX and BAK did not correlate with cell survival. Using absolute protein levels as input for DR_MOMP, we found a strongly improved correlation between model predictions and cells’ responses to Cisplatin (ρ = 0.93) and Paclitaxel treatments (ρ = 0.97). Next we performed synergy studies using the selective BCL2 antagonist ABT-199 or the selective BCL(X)L antagonist WEHI-539 in combination with cisplatin. We observed differential effects of BCL(X)L and BCL-2 inhibition in individual cell lines that were remodelled by DR_MOMP. Modelling also suggested additional effects of BCL2 antagonists independent of MOMP regulation. BCL2 profiling was also performed in primary tumours of 16 TBNC patients in order to predict the patients’ risk prospectively. DR_MOMP predicted a high risk in 37.5% of patients which is in line with 5-year recurrence rates in the literature [4]. Our findings provide evidence that DR_MOMP can be deployed to predict the response of TNBC breast cancer to genotoxic therapy, and can aid in the choice of the optimal BCL2 antagonists. This work is supported by Irish Cancer Society Collaborative Cancer Research Centre BREAST-PREDICT [1] Pillai et al. Am J Cancer Res. 2013 Jun 20;3(3):312-22. PubMed PMID: 23841030. [2] Lee et al. BMC Cancer. 2007 Apr 12;7:63. PubMed PMID: 17430582. [3] Lindner et al. Cancer Res. 2013 Jan 15;73(2):519-28. PubMed PMID: 23329644 [4] Haffty et al. J Clin Oncol. 2006 Dec 20;24(36):5652-7. PubMed PMID: 17116942 Citation Format: Federico Lucantoni, Andreas U. Lindner, Norma O’Donovan, Damir Vareslija, Lance Hudson, Arnold DK Hill, Michael Kerin, Roisin Dwyer, Leonie Young, Jochen HM Prehn. System-based BCL2 family protein signatures as predictive biomarkers in triple-negative breast cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 3557.