Recognition that oral absorption of drugs can be related to molecular structural features and intrinsic physical chemistry properties, namely the Lipinski ‘Rule of Five’ and subsequent ‘drug-likeness’ variants, have provided powerful models in drug discovery for triaging design ideas. Oral drugs that have advanced to clinical development and regulatory approval, despite one or more violations of traditional predictive guidelines, have raised doubts and questions regarding the usefulness of these models in drug discovery. To see if trends could be observed across Ro5 violations in orally absorbed drugs, we experimentally examined Structure-Property Relationships of clinical candidates and marketed drugs. We measured free ligand solution conformations and dynamics for a set of 15 bRo5 compounds. In addition, we explored the experimental ‘drug-likeness’ space using experimental structural descriptors for hydrogen bond donor and acceptor counts, and number of rotatable bonds, as well as chromatographic measurements of lipophilicity (chromLogD) and exposed polar surface area (ePSA). The resultant hypothesis was tested across a set of approximately 100 compounds providing important insights leading to a more accurate prediction of physicochemical property drivers for oral absorption.

RIPK1 kinase inhibition has emerged as a promising therapeutic target for the treatment of neurodegenerative, autoimmune, and inflammatory diseases, including Alzheimer’s disease (AD), amyotrophic lateral sclerosis (ALS) and multiple sclerosis (MS). Extensive genetic studies demonstrate that RIPK1 is involved in tumor necrosis factor (TNF) receptor-1 pathway signaling, thus implicating this target in both (neuro)inflammation and cell death. As such, we sought to identify a potent, selective, low human dose, CNS-penetrant type-III RIPK1 inhibitor for neurodegeneration.

Our strategies consisted of knowledge-based drug design and focused 3D-structure based-virtual screens leveraging our internal compound collection. This effort resulted in the rapid identification of several novel promising hits including compound 1. A focused lead identification campaign utilizing structure based drug design and QSAR prediction models focusing on CNS properties and pharmacokinetics led to the discovery of highly potent and selective lead series with good cross-species pharmacokinetic properties and without Pgp and BCRP-transporter liabilities. We utilized an adapted research operating plan to accelerate the discovery of advanced lead compound 2 a type-III RIPK1 inhibitor with favorable pharmacokinetic profile in pre-clinical species and excellent selectivity.

The heterodimeric transmembrane receptor, integrin avb6 is a promising potential target for the treatment of idiopathic pulmonary fibrosis (IPF) and other fibrotic diseases. In this talk, we will describe the discovery of MORF-627, an orally bioavailable, selective avb6 integrin antagonist with activity in preclinical models of IPF. Optimization of early non-selective hits was accomplished using our MiNT platform, which integrates broad in vitro assay profiling, high-resolution X-ray co-crystal structure determination and in silico potency prediction using Free Energy Perturbation (FEP), to deliver a candidate with potency and pharmacokinetic (PK) properties that support a low projected human dose.

Targeted protein degradation has emerged in recent years as a new modality to control protein levels in vivo. Among the many competing degradation technologies, most research to date has focused on heterobifunctional protein degraders, although inherent concerns about the molecular properties of these “large small molecules” has sparked intense interest in the development of smaller degraders, such as molecular glues. As a rule, molecular glues are smaller and are thus more attractive as potential therapeutics – but because a single molecule is responsible for simultaneously binding to two disparate proteins, the rational discovery of molecular glues to date has proven difficult. In this work, we will discuss multiple computational techniques implemented in MOE for modeling ternary complexes containing molecular glues. Special focus will be paid to lessons learned from our well-established bifunctional modeling toolset. In particular, it will be shown that the most effective molecular glue models result from treating molecular glues as “linkerless bifunctionals.” Finally, recent work in developing techniques to prioritize prospective molecular glue designs based on their predicted degradation efficacy will be presented.

cGMP-dependent protein kinase I-α (PKG1α) is a target for pulmonary arterial hypertension due to its role in the regulation of smooth muscle function. While most work has focused on regulation of cGMP turnover, we recently described several small molecule tool compounds which were capable of activating PKG1α via a cGMP independent pathway. Selected molecules were crystallized in the presence of PKG1α and were found to bind to an allosteric site proximal to the low-affinity nucleotide binding domain. These molecules act to displace the regulatory helix and cause activation of PKG1α representing a new mechanism for the activation and control of this critical therapeutic path. The described structures are vital to understanding the function and control of this key regulatory pathway.

Transcriptional dysregulation is a hallmark of many cancers (PMID: 28187285) and is exemplified by genomic amplifications to the MYC family of oncogenes which occur in at least 20% of all adult solid tumors (PMID: 29596783). Targeting of transcriptional cofactors and in particular the transcriptional cyclin dependent kinase, CDK9, has emerged as a therapeutic strategy to interdict deregulated transcriptional activity including oncogenic MYC. Here, we report the structural optimization of a small molecule microarray hit, prioritizing maintenance of CDK9 selectivity while improving on-target potency and overall physicochemical and pharmacokinetic properties. This led to the discovery of the potent, selective, orally bioavailable CDK9 inhibitor KB-0742. KB-0742 exhibits in vivo antitumor activity in mouse xenograft models, a toxicology profile supporting clinical dosing, and a projected human PK profile anticipated to enable efficacious oral dosing. KB-0742 is currently in a Phase 1/2 dose escalation and expansion clinical trial in patients with relapsed or refractory solid tumors (NCT04718675).