Abstract AEG1 is highly expressed in solid tumors; hyperexpression confers multidrug resistance, increased tumor proliferation, angiogenesis, metastatic potential and poor outcome in most solid tumors. AEG1 activates multiple protumorogenic signal transduction pathways like PIK3/ALT, NFKB, MAMPK, WNT and is a rational target for therapeutic intervention in solid tumors.Also AEG1 functionality is instrumental in the natural history of brain tumors. Our data in NSCLCs confirm AEG1 overexpression as the main parameter (uni and multivariate analysis) indicative of outcome to targeted and cytotoxic therapy. We designed an in silico model based on inhibition of AEG1-p65 crosstalk, identifying a possible interacting interface between the 2 proteins that could be modeled as a three-dimensional structure. This interface was further reduced and transformed into a 3-point pharmacophore for virtual screening comparing both molecular interaction fields and the pharmacophore. A total of 217.000 chemical entities (CEs) were tested in this model giving a Hercules® affinity ranking. Sixty CEs with the highest Hercules® ranking were identified for a second (16 CEs) and third (5 CEs) selection to implement the first round of in vitro studies in our human solid tumor panel. Twenty analogs were generated to seek for a lead according to prospective selection criteria. PB0412-3 (PB3), a small molecule polyheterocyclic compound, was selected for further development. PB3 displays antitumor activity in the human solid tumor panel (median IC50 1.3 uM) and PB3-induced growth inhibition appears independent of EGFR, KRAS & p53 mutational status, Her2 amplification, PIK3 mutations and BRCA1 expression;tumor cell lines bearing highest AEG1-mRNA expression levels were sensititive to PB3. Median IC50 in our glioblastoma (GBM) panel was 60 nM with no impact of MGMT methylation status on PB3's molar potency (MP). IC50s delta solid vs brain tumors were significant (p<.0001) representing a 25-fold increase in MP. Data from nitrosureas in the same GBM panel indicate PB3 is 250 times more potent than BCNU and temozolamide. Median IC50s in human non cancer lines is 2 uM. PB3 was re-screened in the in silico model; PB3 fully matches the AEG1 binder pharmacophore model and was screened in the Chemical Space web tool; results confirm PB3 as first-in-class CE. Functional studies are ongoing; preliminary data in PC9 NSCLC cell line demonstrate significant PB3-induced downregulation of AEG1 mRNA expression. Expansion of the CNS tumor panel, cell kinetics, drug-induced cell death & in vivo/PK-PD studies in SC orthotopic models is ongoing. To the best of our knowledge, this is the first successful attempt at the Pharmacological intervention against the AEG1 pathway. A dual developmental plan for PB3 as local brain drug delivery and systemic therapy is proposed. Citation Format: Jose M. Jimeno, Gerardo Acosta, Miguel Ángel Molina, Nicky Karachaliou, Cristina Teixidó, Carlos Obiol, Oriol Villacañas, Jordi Bertran, María Sánchez Rouco, Ana Giménez Capitán, Santiago Ramón y Cajal, Miquel Taron, Rafael Rosell, Fernando Albericio. Astrocytic elevated gene 1 (AEG1) a target for pharmacological anticancer intervention. [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 4601. doi:10.1158/1538-7445.AM2014-4601
Read full abstract