Targeting Aurora Kinase in Aggressive B-Cell Non-Hodgkin's Lymphomas.

Abstract 284

Aurora kinases (A and B) are oncogenic serine/threonine (S/T) kinases that play central roles in the mitotic phase of the eukaryotic cell cycle. Over-expression of Aurora kinases during the cell cycle over-rides mitotic and spindle check points leading to aneuploidy in many human cancers; Aurora kinases are therefore attractive therapeutic targets. Gene expression profiling in aggressive B- and T-cell non-Hodgkin's lymphoma (NHL) has shown the Aurora kinases to be over-expressed and they may be key component genes of the ‘proliferative' signature. We hypothesized (1) Aurora kinases are over-expressed in human aggressive B-cell NHL (mantle cell lymphoma (MCL), diffuse large B-cell lymphoma (DLBCL) and transformed follicular lymphoma (TFL)), (2) Aurora ATP-binding site small molecule inhibitor (SMI) is effective in promoting apoptosis in cell culture and tumor growth inhibition (TGI) in mouse xenograft model(s) of NHL, and (3) Aurora SMI will be safe and effective in treating patients with relapsed aggressive B-cell NHL in early phase clinical trials.

To analyze Aurora expression, tissue microarrays (TMA) were constructed from 43 patients with DLBCL and 40 with MCL, and Aurora A and B expression optimized with commercially available antibodies by immunohistochemistry (IHC). The IHC was rated as a staining intensity on a scale of 0 to 3+. The NHL TMAs demonstrated intense staining (2+ to 3+) for Aurora A (nucleus) and Aurora B (nucleus) in >60% compared to normal lymph nodes. We also analyzed the Lymphoma/Leukemia Molecular Profiling Project (LLMPP) publicly available database for Aurora A and B expression in MCL and found a worse survival in those with A > B over-expression (p<0.01). Since both Auroras are transforming genes, the LLMPP data support the conclusion that these S/T kinases are associated with a poor prognosis and are potential targets for therapy.

Western blotting analysis of 13 B-cell NHL cell lines (DLBCL, MCL and TFL) for Aurora A and B expression showed significant over-expression compared to B-cells isolated from normal lymph nodes. Aurora A knockdown by shRNA in the B-NHL cell lines showed inhibition of mitosis with a polyploid phenotype (4n, 8n) that ends in apoptosis as shown by PARP-cleavage. The Aurora A specific inhibitor (MLN8237) evaluated in the 13 NHL cell lines phenocopies shRNA knockdown with associated inhibition of proliferation (IC50=0.05 mM) and promotes apoptosis (flow cytometry, PARP-cleavage) in a dose-dependent manner. Combination of MLN8237 with a microtubule targeting agent (docetaxel) to abrogate the spindle checkpoint is synergistic in a sequence specific manner. Moreover, MLN8237 effectively inhibits Aurora A auto-phosphorylation and eliminates phospho-histone H3 (Ser10) phosphorylation. Currently, 2 mouse MCL (Granta 519) xenograft models are underway evaluating tumor growth inhibition (TGI), safety and survival of MLN8237 alone and in combination with docetaxel or rituximab respectively. Preliminary data show that MLN8237 is synergistic with docetaxel in TGI in the MCL xenograft model. Together the data suggest inhibition of Aurora kinases may offer a promising treatment strategy for patients with aggressive B-cell NHL [Funded by the Lymphoma SPORE, P50 CA130805501A1, PI: Richard Fisher, MD.].