Abstract 5449A: PI3K and MEK inhibition in intracranial triple negative breast cancer: Efficacy of BKM120 and AZD6244 in preclinical mouse models

Introduction: Triple-negative breast cancer (TNBC) is an aggressive subtype which lacks the classical clinical biomarkers for hormone receptors (HR) and HER2. Nearly half of metastatic TNBC patients develop brain metastases and these patients face a poor prognosis. Most new effective treatments for breast cancer target hyperactivated HER2 and HR pathways, which TNBC tumors lack. The blood-brain barrier (BBB) also prevents many treatments from reaching the brain and any developing tumors therein. There is currently no FDA-approved systemic chemotherapy for the treatment of TNBC brain metastases. Tissue-based studies show activation of the PI3K pathway in breast cancer brain metastases and in vivo models of extracranial TNBC are sensitive to PI3K and MEK inhibitors. BBB-permeable inhibitors targeting the PI3K and MEK pathways are in clinical development. In this study, we evaluated the efficacy of two such inhibitors, the pan-PI3K inhibitor BKM120 and the MEK1/2 inhibitor AZD6244, in preclinical models of intracranial (IC) TNBC. Methods: The efficacy of PI3K and MEK inhibition by BKM120 and AZD6244, respectively, alone and in combination was determined in orthotopic mouse models of basal-like TNBC through IC implantation of SUM149 or MDA-MB-468 cell lines. Drugs were administered at the maximum tolerated doses via chow: 30 mg/kg BKM, 37 mg/kg AZD, or 25 mg/kg BKM + 18 mg/kg AZD. Tumor burden was monitored via weekly bioluminescence imaging, and brain tumors were frozen at sacrifice for gene expression analyses. To explore potential resistance mechanisms and inform rational combination therapies, in vitro IC50s, combination synergy determinations, and single-agent synthetic lethal siRNA screens were conducted. Results: In the SUM149 model, median survival was 45 days (control), 53.5 days (BKM), 53.5 days (BKM+AZD), and 76 days (AZD). AZD alone and with BKM reduced tumor burden via bioluminescence imaging. In vitro, 3-day dose response curves showed that BKM and AZD were similarly potent but BKM was more effective (IC50; maximal % cell reduction: BKM: 1.3 uM, >90%; AZD: 816 nM, >50%). Several genes were identified as synthetically lethal “hits” in an initial kinome screen by two-class SAM. AURKA was a hit with BKM120, while BRAF and several PI3K, AMPK, and CDK genes were synthetically lethal with AZD6244. Ongoing work will include survival studies in the 468 model and gene expression changes in IC tumors in response to therapy. Conclusions: BKM120 and AZD6244 both improved survival in an IC TNBC SUM149 mouse model, with single agent AZD6244 being most efficacious. The siRNA screens indicate that combined treatment with BKM120 and AZD6244 should be synthetically lethal, suggesting that combination therapy may have underperformed due to toxicity. Ongoing in vitro and in vivo studies (including dosing schedules) will further characterize the effects of these drugs in intracranial TNBC. Citation Format: Amanda E.D. Van Swearingen, Marni B. Siegel, Ryan Bash, Brian Golitz, Charlene Santos, David Darr, Joel Parker, Gary L. Johnson, C. Ryan Miller, Carey K. Anders. PI3K and MEK inhibition in intracranial triple negative breast cancer: Efficacy of BKM120 and AZD6244 in preclinical mouse models. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 5449A. doi:10.1158/1538-7445.AM2014-5449A