Erik Kupperman

Evaluation of the Proteasome Inhibitor MLN9708 in Preclinical Models of Human Cancer

The proteasome was validated as an oncology target following the clinical success of VELCADE (bortezomib) for injection for the treatment of multiple myeloma and recurring mantle cell lymphoma. Consequently, several groups are pursuing the development of additional small-molecule proteasome inhibitors for both hematologic and solid tumor indications. Here, we describe MLN9708, a selective, orally bioavailable, second-generation proteasome inhibitor that is in phase I clinical development. MLN9708 has a shorter proteasome dissociation half-life and improved pharmacokinetics, pharmacodynamics, and antitumor activity compared with bortezomib. MLN9708 has a larger blood volume distribution at steady state, and analysis of 20S proteasome inhibition and markers of the unfolded protein response confirmed that MLN9708 has greater pharmacodynamic effects in tissues than bortezomib. MLN9708 showed activity in both solid tumor and hematologic preclinical xenograft models, and we found a correlation between greater pharmacodynamic responses and improved antitumor activity. Moreover, antitumor activity was shown via multiple dosing routes, including oral gavage. Taken together, these data support the clinical development of MLN9708 for both hematologic and solid tumor indications.

Antitumor activity of the investigational proteasome inhibitor MLN9708 in mouse models of B-cell and plasma cell malignancies.

Purpose: The clinical success of the first-in-class proteasome inhibitor bortezomib (VELCADE) has validated the proteasome as a therapeutic target for treating human cancers. MLN9708 is an investigational proteasome inhibitor that, compared with bortezomib, has improved pharmacokinetics, pharmacodynamics, and antitumor activity in preclinical studies. Here, we focused on evaluating the in vivo activity of MLN2238 (the biologically active form of MLN9708) in a variety of mouse models of hematologic malignancies, including tumor xenograft models derived from a human lymphoma cell line and primary human lymphoma tissue, and genetically engineered mouse (GEM) models of plasma cell malignancies (PCM). Experimental Design: Both cell line–derived OCI-Ly10 and primary human lymphoma–derived PHTX22L xenograft models of diffuse large B-cell lymphoma were used to evaluate the pharmacodynamics and antitumor effects of MLN2238 and bortezomib. The iMycCα/Bcl-XL GEM model was used to assess their effects on de novo PCM and overall survival. The newly developed DP54-Luc–disseminated model of iMycCα/Bcl-XL was used to determine antitumor activity and effects on osteolytic bone disease. Results: MLN2238 has an improved pharmacodynamic profile and antitumor activity compared with bortezomib in both OCI-Ly10 and PHTX22L models. Although both MLN2238 and bortezomib prolonged overall survival, reduced splenomegaly, and attenuated IgG2a levels in the iMycCα/Bcl-XL GEM model, only MLN2238 alleviated osteolytic bone disease in the DP54-Luc model. Conclusions: Our results clearly showed the antitumor activity of MLN2238 in a variety of mouse models of B-cell lymphoma and PCM, supporting its clinical development. MLN9708 is being evaluated in multiple phase I and I/II trials. Clin Cancer Res; 17(23); 7313–23. ©2011 AACR.

TAK-960, a Novel, Orally Available, Selective Inhibitor of Polo-Like Kinase 1, Shows Broad-spectrum Preclinical Antitumor Activity in Multiple Dosing Regimens

Polo-like kinase 1 (PLK1) is a serine/threonine protein kinase involved in key processes during mitosis. Human PLK1 has been shown to be overexpressed in various human cancers, and elevated levels of PLK1 have been associated with poor prognosis, making it an attractive target for anticancer therapy. TAK-960 [4-[(9-cyclopentyl-7,7-difluoro-5-methyl-6-oxo-6,7,8,9-tetrahydro-5H-pyrimido[4,5-b][1,4]diazepin-2-yl)amino]-2-fluoro-5-methoxy-N-(1-methylpiperidin-4-yl) benzamide] is a novel, investigational, orally bioavailable, potent, and selective PLK1 inhibitor that has shown activity in several tumor cell lines, including those that express multidrug-resistant protein 1 (MDR1). Consistent with PLK1 inhibition, TAK-960 treatment caused accumulation of G2–M cells, aberrant polo mitosis morphology, and increased phosphorylation of histone H3 (pHH3) in vitro and in vivo. TAK-960 inhibited proliferation of multiple cancer cell lines, with mean EC50 values ranging from 8.4 to 46.9 nmol/L, but not in nondividing normal cells (EC50 >1,000 nmol/L). The mutation status of TP53 or KRAS and MDR1 expression did not correlate with the potency of TAK-960 in the cell lines tested. In animal models, oral administration of TAK-960 increased pHH3 in a dose-dependent manner and significantly inhibited the growth of HT-29 colorectal cancer xenografts. Treatment with once daily TAK-960 exhibited significant efficacy against multiple tumor xenografts, including an adriamycin/paclitaxel-resistant xenograft model and a disseminated leukemia model. TAK-960 has entered clinical evaluation in patients with advanced cancers. Mol Cancer Ther; 11(3); 700–9. ©2011 AACR.

Abstract A207: Discovery of molecular mechanisms underlying cellular sensitivity to the PLK1 inhibitor TAK-960.

Background: Polo-like kinase 1 (PLK1) plays an essential role in mitosis. Human PLK1 has been shown to be overexpressed in various human cancers, and has been associated with poor prognosis. TAK-960 is a highly selective, orally bioavailable PLK1 inhibitor that inhibits proliferation in multiple cancer cell lines; however, some cancer cell lines are insensitive to TAK-960. To investigate mechanisms of sensitivity to TAK-960 treatment and to identify a potential gene signature that might be useful for clinical development, we examined gene expression, mutation status, and copy number information in cancer cell lines with different TAK-960 sensitivities. Methods: A total of 100 cancer cell lines were treated with 15–5,000 nmol/L of TAK-960 at ONCOTEST GmbH (Freiburg, Germany) using a clonogenic assay. Cell lines with EC50 values 1,000 nmol/L were defined as sensitive and resistant cell lines, respectively. Gene expression data generated using DNA microarray were obtained from ONCOTEST and used to detect sensitive and resistant cell lines which were then analyzed using Ingenuity Pathway Analysis (Ingenuity Systems, Inc., Redwood City, CA) and Molecular Concepts Analysis (Rhodes D, et al Neoplasia 2007). In vivo antitumor activity of TAK-960 in 15 mouse xenograft models was evaluated by the treated/control (T/C) tumor volume ratio after once daily oral administration of TAK-960 10 mg/kg for 2 weeks. Mutation status and copy number information of cell lines used to establish the xenograft models were obtained from the Catalogue of Somatic Mutations in Cancer (COSMIC) database. Results: Analysis of gene expression data from ONCOTEST cancer cell lines revealed that expression of several genes implicated in the cell-to-cell signaling and interaction, hematological system development and function, hepatic system disease, cell death, tumor morphology, and cellular development network including CDKN2A, CD36, TLR4, TNFRSF11A, IGFBP1, TIMP3 and several chemokine genes were different in sensitive and resistant cell lines. In addition, xenograft tumors (A549, MES-SA/D×5, 786-O, Caki-1, and Ma-1) with a deletion or mutation of CDKN2A, which is a tumor suppressor gene encoding the p16 cell cycle regulatory protein, did not respond to TAK-960 compared with CDKN2A wild-type xenograft tumors (BT474, HT-29, MV4–11, PC-3, H1299, and A2780). Conclusions: These results indicate that CDKN2A status correlated with sensitivity and thus suggest that CDKN2A status may be a useful predictive biomarker of TAK-960 antitumor activity in clinical studies. The role of CDKN2A in modulation of sensitivity to TAK-960 is under investigation. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2011 Nov 12-16; San Francisco, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2011;10(11 Suppl):Abstract nr A207.

Second-Generation Proteasome Inhibitors

The first-in-class proteasome inhibitor, bortezomib, has provided proof-of-concept for the therapeutic approach of proteasome inhibition in a number of malignancies. However, as we look to the future and to further improving upon the contributions of this class of drugs, we will need to consider optimizing activity in solid tumors, reducing peripheral neuropathy and utilizing more convenient routes of administration. A number of “second-generation” proteasome inhibitors have been identified and are now in preclinical and clinical development, including MLN9708, CEP-18770, carfilzomib, and salinosporamide A (NPI-0052). These agents differ from bortezomib in some of their key characteristics, and differences in their pharmacology may result in different activity and safety profiles. This chapter reviews the second-generation proteasome inhibitors, together with other potential therapeutic targets in the ubiquitin–proteasome system.

Abstract C99: Beyond bortezomib: Development of Millennium's next-generation proteasome inhibitors.

Bortezomib (Velcade®) is a proteasome inhibitor that has been approved by the U.S. Food and Drug Administration for the treatment of patients with multiple myeloma or relapsed mantle cell lymphoma. It is currently available in more than 90 countries worldwide. The molecule is an N-capped dipeptidyl boronic acid and its molecular mechanism involves slow-tight binding to the chymotrypsin-like (β5) sites of the 26S proteasome. In the development of Millennium9s next generation inhibitors, we have investigated several parameters affecting the drug9s activity in biological systems. These include tissue proteasome concentration, blood/plasma partitioning and the kinetics of proteasome inhibition in cultured cells. We find that the abundance of the proteasome (approx. 1–5 μM β5 active site concentration) in cells and tissues together with the slow rate of dissociation of bortezomib from the proteasome (110 min. half-life) led to partitioning of the inhibitor in red blood cells, thereby limiting its distribution to potential sites of therapeutic action. Based on the hypothesis that more rapid dissociation from the proteasome should improve tissue distribution, we have developed an investigational, dipeptidyl boronic acid proteasome inhibitor (MLN9708). MLN9708 displays comparable potency (i.e. similar Ki) to bortezomib for the β5 site of the proteasome but a shorter (18 min.) half-life of dissociation. This affects rapid recovery of proteasome activity in tissue culture cells upon washout of the drug as well as reduced blood/plasma partitioning in mice, supporting the hypothesis that a more rapid equilibrium proteasome inhibitor can improve tissue distribution. MLN9708 is currently in phase 1 clinical trials. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2011 Nov 12-16; San Francisco, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2011;10(11 Suppl):Abstract nr C99.

Abstract B40: Development of a pharmacodynamic assay for the second‐generation proteasome inhibitor MLN9708 in clinical trials of multiple myeloma

The ubiquitin‐proteasome system processes the majority of cellular proteins and is the principal pathway by which cells regulate protein homeostasis. The successful development of VELCADE® (bortezomib) for Injection for multiple myeloma and previously treated mantle cell lymphoma has validated the proteasome as a therapeutic target for the treatment of malignancies. MLN9708 is a second‐generation reversible proteasome inhibitor developed to achieve greater oral bioavailability, improved pharmacokinetics and greater antitumor activity than bortezomib. MLN9708 is currently in human clinical development for both hematological and non‐hematological malignancies. MLN9708 immediately hydrolyzes to the biologically active form MLN2238 upon exposure to aqueous solutions or plasma, and MLN2238 was used for all preclinical studies described below. In vitro, MLN2238 inhibited 20S proteasome activity, preferentially binding the 20S 5 site with an IC50 of 3.4 nM and demonstrated potent activity against cultured cancer cells in cell viability assays. In vivo, MLN2238 achieved exposures that resulted in significant blood and tumor proteasome inhibition in xenograft‐bearing mice and had increased plasma and tumor exposure compared to bortezomib when dosed at their respective maximum tolerated doses (MTD). MLN2238 also elicited a stronger pharmacodynamic (PD) response than bortezomib in xenograft tumors, as measured by tumor 20S 5 site‐specific activity and expression levels of GADD34 and ATF‐3, two genes involved in the unfolded protein response (UPR) pathway shown to be upregulated in response to proteasome inhibition. Here we describe the development of an ATF‐3 IHC assay suitable for use in bone marrow samples isolated from multiple myeloma patients enrolled in Phase I trials of MLN9708. Using both quantitativeWestern blotting and IHC assays, we demonstrated that treatment of cultured myeloma cell lines with MLN2238 results in a dose‐dependent increase in ATF‐3 levels. Antibody specificity was confirmed by IHC analysis of HCT‐116 cells following knockdown of ATF‐3 by siRNA. An ATF‐3 PD response is detectable by IHC in several xenograft models and in selected normal tissues from mice dosed with MLN2238. The elevation in ATF‐3 is delayed by several hours compared to 20S proteasome inhibition in cells and tissues, consistent with it being a downstream effect of proteasome inhibition. To test the use of the ATF‐3 IHC assay on clinical samples from multiple myeloma patients, we developed a dual staining assay with CD38, a marker expressed on multiple myeloma cells. We show baseline staining of ATF‐3 and CD38 in formalin‐fixed paraffin embedded bone marrow aspirates and biopsies from patients who have not received MLN9708. This assay has potential for use in evaluating the levels of CD38 and ATF‐3 in pre‐ and post‐treatment bone marrow samples from patients treated with MLN9708. Citation Information: Mol Cancer Ther 2009;8(12 Suppl):B40.

Perspectives on molecular and anatomical imaging in drug discovery

INTRODUCTION The role of medical imaging in the preclinical, pharmaceutical environment has entered a growth phase in the drug discovery setting due to at least three main factors: (i) improved technology, specific to the imaging of small animals; (ii) the high-visibility impact of imaging on the practice of human medicine; and (iii) operational advantages leading to a competitive advantage over other drug discovery organizations (1,2). However, it is important to note that as much as one could make the case to employ imaging, it must also be emphasized that an organizational investment—both monetary and operationally—must be made for the proper implementation and integration of imaging into a company’s drug development work flow (1,2). Still the question remains: What is the role of imaging, both molecular and anatomical, in improving the current drug discovery process? This article examines this question in light of the experiences of the authors in establishing an Imaging Sciences Group at Millennium Pharmaceuticals (Cambridge, MA, USA), as well as personal communications with colleagues in the field. First, we will briefly review why imaging could impact pharmaceutical drug development—a process wrought with potential failures—and then highlight specific examples of the approaches that we have taken.