Discovery of the c-Jun N-Terminal Kinase Inhibitor CC-90001.

As a result of emerging biological data suggesting that within the c-Jun N-terminal kinase (JNK) family, JNK1 and not JNK2 or JNK3 may be primarily responsible for fibrosis pathology, we sought to identify JNK inhibitors with an increased JNK1 bias relative to our previous clinical compound tanzisertib (CC-930). This manuscript reports the synthesis and structure-activity relationship (SAR) studies for a novel series of JNK inhibitors demonstrating an increased JNK1 bias. SAR optimization on a series of 2,4-dialkylamino-pyrimidine-5-carboxamides resulted in the identification of compounds possessing low nanomolar JNK inhibitory potency, overall kinome selectivity, and the ability to inhibit cellular phosphorylation of the direct JNK substrate c-Jun. Optimization of physicochemical properties in this series resulted in compounds that demonstrated excellent systemic exposure following oral dosing, enabling in vivo efficacy studies and the selection of a candidate for clinical development, CC-90001, which is currently in clinical trials (Phase II) in patients with idiopathic pulmonary fibrosis (NCT03142191).

Structure-Guided Optimization Provides a Series of TTK Protein Inhibitors with Potent Antitumor Activity.

TTK is an essential spindle assembly checkpoint enzyme in many organisms. It plays a central role in tumor cell proliferation and is aberrantly overexpressed in a wide range of tumor types. We recently reported on a series of potent and selective TTK inhibitors with strong antiproliferative activity in triple negative breast cancer (TNBC) cell lines (8: TTK IC50 = 3.0 nM; CAL-51 IC50 = 84.0 nM). Inspired by previously described potent tricyclic TTK inhibitor 6 (TTK IC50 = 0.9 nM), we embarked on a structure-enabled design and optimization campaign to identify an improved series with excellent potency, TTK selectivity, solubility, CYP inhibition profile, and in vivo efficacy in a TNBC xenograft model. These efforts culminated in the discovery of 25 (TTK IC50 = 3.0 nM; CAL-51 IC50 = 16.0 nM), which showed significant single-agent efficacy when dosed iv in a TNBC xenograft model without body weight loss.

Discovery of the Selective Protein Kinase C-θ Kinase Inhibitor, CC-90005.

The PKC-θ isoform of protein kinase C is selectively expressed in T lymphocytes and plays an important role in the T cell antigen receptor (TCR)-triggered activation of mature T cells, T cell proliferation, and the subsequent release of cytokines such as interleukin-2 (IL-2). Herein, we report the synthesis and structure-activity relationship (SAR) of a novel series of PKC-θ inhibitors. Through a combination of structure-guided design and exploratory SAR, suitable replacements for the basic C4 amine of the original lead (3) were identified. Property-guided design enabled the identification of appropriately substituted C2 groups to afford potent analogs with metabolic stability and permeability to support in vivo testing. With exquisite general kinase selectivity, cellular inhibition of T cell activation as assessed by IL-2 expression, a favorable safety profile, and demonstrated in vivo efficacy in models of acute and chronic T cell activation with oral dosing, CC-90005 (57) was selected for clinical development.

The challenges of generalizability in artificial intelligence for ADME/Tox endpoint and activity prediction.

INTRODUCTION: Artificial intelligence (AI) has seen a massive resurgence in recent years with wide successes in computer vision, natural language processing, and games. The similar creation of robust and accurate AI models for ADME/Tox endpoint and activity prediction would be revolutionary to drug discovery pipelines. There have been numerous demonstrations of successful applications, but a key challenge remains: how generalizable are these predictive models? AREAS COVERED: The authors present a summary of current promising components of AI models in the context of early drug discovery where ADME/Tox endpoint and activity prediction is the main driver of the iterative drug design process. Following that is a review of applicability domains and dataset construction considerations which determine generalizability bottlenecks for AI deployment. Further reviewed is the role of promising learning frameworks - multitask, transfer, and meta learning - which leverage auxiliary data to overcome issues of generalizability. EXPERT OPINION: The authors conclude that the most promising direction toward integrating reliable and informative AI models into the drug discovery pipeline is a conjunction of learned feature representations, deep learning, and novel learning frameworks. Such a solution would address the sparse and incomplete datasets that are available for key endpoints related to drug discovery.

Design and Optimization Leading to an Orally Active TTK Protein Kinase Inhibitor with Robust Single Agent Efficacy.

Triple negative breast cancer (TNBC) is an aggressive disease with high relapse rates and few treatment options. Outlined in previous publications, we identified a series of potent, dual TTK/CLK2 inhibitors with strong efficacy in TNBC xenograft models. Pharmacokinetic properties and kinome selectivity were optimized, resulting in the identification of a new series of potent, selective, and orally bioavailable TTK inhibitors. We describe here the structure-activity relationship of the 2,4-disubstituted-7 H-pyrrolo[2,3- d]pyrimidine series, leading to significant single agent efficacy in a TNBC xenograft model without body weight loss. The design effort evolving an iv-dosed TTK/CLK2 inhibitor to an orally bioavailable TTK inhibitor is described.

The Discovery of a Dual TTK Protein Kinase/CDC2-Like Kinase (CLK2) Inhibitor for the Treatment of Triple Negative Breast Cancer Initiated from a Phenotypic Screen.

Triple negative breast cancer (TNBC) remains a serious unmet medical need with discouragingly high relapse rates. We report here the synthesis and structure-activity relationship (SAR) of a novel series of 2,4,5-trisubstituted-7H-pyrrolo[2,3-d]pyrimidines with potent activity against TNBC tumor cell lines. These compounds were discovered from a TNBC phenotypic screen and possess a unique dual inhibition profile targeting TTK (mitotic exit) and CLK2 (mRNA splicing). Design and optimization, driven with a TNBC tumor cell assay, identified potent and selective compounds with favorable in vitro and in vivo activity profiles and good iv PK properties. This cell-based driven SAR produced compounds with strong single agent in vivo efficacy in multiple TNBC xenograft models without significant body weight loss. These data supported the nomination of CC-671 into IND-enabling studies as a single agent TNBC therapy.

Discovery of mammalian target of rapamycin (mTOR) kinase inhibitor CC-223.

We report here the synthesis and structure-activity relationship (SAR) of a novel series of mammalian target of rapamycin (mTOR) kinase inhibitors. A series of 4,6- or 1,7-disubstituted-3,4-dihydropyrazino[2,3-b]pyrazine-2(1H)-ones were optimized for in vivo efficacy. These efforts resulted in the identification of compounds with excellent mTOR kinase inhibitory potency, with exquisite kinase selectivity over the related lipid kinase PI3K. The improved PK properties of this series allowed for exploration of in vivo efficacy and ultimately the selection of CC-223 for clinical development.

Optimization of a Series of Triazole Containing Mammalian Target of Rapamycin (mTOR) Kinase Inhibitors and the Discovery of CC-115.

We report here the synthesis and structure-activity relationship (SAR) of a novel series of triazole containing mammalian target of rapamycin (mTOR) kinase inhibitors. SAR studies examining the potency, selectivity, and PK parameters for a series of triazole containing 4,6- or 1,7-disubstituted-3,4-dihydropyrazino[2,3-b]pyrazine-2(1H)-ones resulted in the identification of triazole containing mTOR kinase inhibitors with improved PK properties. Potent compounds from this series were found to block both mTORC1(pS6) and mTORC2(pAktS473) signaling in PC-3 cancer cells, in vitro and in vivo. When assessed in efficacy models, analogs exhibited dose-dependent efficacy in tumor xenograft models. This work resulted in the selection of CC-115 for clinical development.

Aminopurine based JNK inhibitors for the prevention of ischemia reperfusion injury.

In this Letter we describe the optimization of an aminopurine lead (1) with modest potency and poor overall kinase selectivity which led to the identification of a series of potent, selective JNK inhibitors. Improvement in kinase selectivity was enabled by introduction of an aliphatic side chain at the C-2 position. CC-359 (2) was selected as a potential clinical candidate for diseases manifested by ischemia reperfusion injury.

Discovery of CC-930, an orally active anti-fibrotic JNK inhibitor.

In this Letter we describe the discovery of potent, selective, and orally active aminopurine JNK inhibitors. Improving the physico-chemical properties as well as increasing the potency and selectivity of a subseries with rat plasma exposure, led to the identification of four structurally diverse inhibitors. Differentiation based on PK profiles in multiple species as well as activity in a chronic efficacy model led to the identification of 1 (CC-930) as a development candidate, which is currently in Phase II clinical trial for IPF.

Fluorination-free synthesis of a 4,4-difluoro-3,3-dimethylproline derivative.

A Claisen rearrangement/iodolactamization sequence starting from commercially available trifluoroacetaldehyde methyl hemiacetal, followed by a classical chemical resolution, provided enantiomerically pure 4,4-difluoro-3,3-dimethylproline (S)-1. No hazardous fluorination reagents were used, and the overall yield over 12 steps was greater than 28%.