Martha A. Sarpong

Discovery of GSK2126458, a Highly Potent Inhibitor of PI3K and the Mammalian Target of Rapamycin.

Phosphoinositide 3-kinase α (PI3Kα) is a critical regulator of cell growth and transformation, and its signaling pathway is the most commonly mutated pathway in human cancers. The mammalian target of rapamycin (mTOR), a class IV PI3K protein kinase, is also a central regulator of cell growth, and mTOR inhibitors are believed to augment the antiproliferative efficacy of PI3K/AKT pathway inhibition. 2,4-Difluoro-N-{2-(methyloxy)-5-[4-(4-pyridazinyl)-6-quinolinyl]-3-pyridinyl}benzenesulfonamide (GSK2126458, 1) has been identified as a highly potent, orally bioavailable inhibitor of PI3Kα and mTOR with in vivo activity in both pharmacodynamic and tumor growth efficacy models. Compound 1 is currently being evaluated in human clinical trials for the treatment of cancer.

Discovery of GSK1070916, a Potent and Selective Inhibitor of Aurora B/C Kinase

The Aurora kinases play critical roles in the regulation of mitosis and are frequently overexpressed or amplified in human tumors. Selective inhibitors may provide a new therapy for the treatment of tumors with Aurora kinase amplification. Herein we describe our lead optimization efforts within a 7-azaindole-based series culminating in the identification of GSK1070916 (17k). Key to the advancement of the series was the introduction of a 2-aryl group containing a basic amine onto the azaindole leading to significantly improved cellular activity. Compound 17k is a potent and selective ATP-competitive inhibitor of Aurora B and C with K(i)* values of 0.38 +/- 0.29 and 1.5 +/- 0.4 nM, respectively, and is >250-fold selective over Aurora A. Biochemical characterization revealed that compound 17k has an extremely slow dissociation half-life from Aurora B (>480 min), distinguishing it from clinical compounds 1 and 2. In vitro treatment of A549 human lung cancer cells with compound 17k results in a potent antiproliferative effect (EC(50) = 7 nM). Intraperitoneal administration of 17k in mice bearing human tumor xenografts leads to inhibition of histone H3 phosphorylation at serine 10 in human colon cancer (Colo205) and tumor regression in human leukemia (HL-60). Compound 17k is being progressed to human clinical trials.

Abstract C62: Identification of GSK2126458, a highly potent inhibitor of phosphoinositide 3‐kinase (PI3K) and the mammalian target of rapamycin (mTOR)

Phosphoinositide 3‐kinase (PI3K) is a critical regulator of cell growth and transformation and its signaling pathway is one of the most commonly mutated pathways in human cancer. The mammalian target of rapamycin (mTOR), a class IV PI3K protein kinase, is also a central regulator of cell growth, and mTOR inhibitors are believed to augment the antiproliferative efficacy of the PI3K/AKT pathway. GSK1059615, our first PI3K clinical compound, progressed to a dose escalation study in patients with refractory malignancies. Following the discovery of GSK1059615, we sought to identify a second inhibitor with improved potency, selectivity, and pharmacokinetics. Key to our approach to achieving the desired levels of PI3K activity was to pursue structure‐based design utilizing crystallography of the more amenable PI3K as a surrogate protein. Following a chemistry lead optimization effort, the pyridylsulfonamide GSK2126458 was identified as a highly potent, orally bioavailable, pan‐PI3K and mTOR inhibitor (PI3K app Ki = 19 pM; mTORC1 app Ki = 180 pM; mTORC2 app Ki = 300 pM). Consistent with potent PI3K and mTORC2 enzyme inhibition, GSK2126458 decreased cellular levels of phosphorylated AKT (BT474 pAKT IC50 = 180 pM) and inhibited cell proliferation in a large panel of cancer cell lines (e.g. BT474 growth IC50 = 2 nM). GSK2126458 showed good exposure in four pre‐clinical animal species and exhibited in vivo activity in both pharmacodynamic and tumor growth efficacy models. GSK2126458 is being evaluated currently in human clinical trials for the treatment of cancer. The discovery, design, and optimization of GSK2126458 and related analogs will be presented. Citation Information: Mol Cancer Ther 2009;8(12 Suppl):C62.

Abstract C74: Reassessing IKKε as a novel oncology target

Recent years have seen increasing efforts to employ integrative genomic approaches to identify novel oncogenes. In this manner, IKBKE coding for IKKe was recently proposed as a novel oncogene, aberrantly activating the NF‐κB signaling pathway in breast cancers (Boehm et al. Cell. 2007). We pursued a series of experiments that delve more deeply into the hypothesis of IKKe as a novel oncology target; the resulting evidence challenges the proposed oncogenic mechanism. We conducted RNA interference studies on mutiple breast cancer cell lines with multiple IKBKE siRNA motifs to define a general on‐target phenotype. Consistent with published findings, we confirmed that although 4 IKBKE siRNA motifs tested significantly reduced IKKe protein expression (78–90% silencing) only 2 of these motifs significantly inhibited cell growth. We then engineered ‘non‐silenceable’ IKBKE constructs packaged in BacMam virus and were able to simultaneously silence the endogenous IKBKE and exogenously express the non‐silenceable IKKe at protein levels below, equivalent to, and significantly above the endogenous levels. We demonstrated that the antiproliferative effects of these 2 siRNAs could not be rescued by functional, exogenously expressed IKBKE, indicating likely off‐target toxicity. We further explored the correlations between IKBKE gene amplification, IKKe protein overexpression, and NF‐κB pathway activation using RNAi in conjunction with an NF‐κB ‐luciferase reporter assay. We observed that IKBKE amplification correlated poorly with both IKKe protein expression and NF‐κB pathway activation. Furthermore, silencing downstream components of the NF‐κB signaling pathway had little impact on the IKBKE‐amplified cells; and conversely, silencing IKBKE had no impact on NF‐κB ‐reporter activity in the IKBKE‐amplified cells. In addition, we over‐expressed the kinase‐dead IKKe (K38A), and observed no effect on tumor cell growth. Finally, using BacMam virus to serially titrate IKK , we confirmed that the level of exogenous IKKe needed to achieve a modest pathway activation (2‐fold increase in NF‐κB ‐luciferase reporter assay) was >10 times the endogenous protein level found in an IKBKE‐amplified breast cancer line. We also undertook a limited medicinal chemistry effort to arrive at several potent small molecule tool inhibitors of both IKK and TBK1 including GSK2292978A (8nM & 1nM respectively). GSK2292978A confirmably inhibited IKKe activity in a cellular mechanistic assay (IC50∼170nM). Using a 3‐day proliferation assay, we observed no evident selectivity of GSK2292978A for IKBKE‐amplified or IKKe overexpressing cell lines compared to non‐amplified cell lines. These findings resulted in our conclusion that IKKe likely does not represent a true oncogene in breast cancer and suggest the vital need for well‐designed controls in cases where RNA interference significantly defines target validation. Citation Information: Mol Cancer Ther 2009;8(12 Suppl):C74.