Discovery of Clinical Candidate AZD0095, a Selective Inhibitor of Monocarboxylate Transporter 4 (MCT4) for Oncology.

Due to increased reliance on glycolysis, which produces lactate, monocarboxylate transporters (MCTs) are often upregulated in cancer. MCT4 is associated with the export of lactic acid from cancer cells under hypoxia, so inhibition of MCT4 may lead to cytotoxic levels of intracellular lactate. In addition, tumor-derived lactate is known to be immunosuppressive, so MCT4 inhibition may be of interest for immuno-oncology. At the outset, no potent and selective MCT4 inhibitors had been reported, but a screen identified a triazolopyrimidine hit, with no close structural analogues. Minor modifications to the triazolopyrimidine were made, alongside design of a constrained linker and broad SAR exploration of the biaryl tail to improve potency, physical properties, PK, and hERG. The resulting clinical candidate 15 (AZD0095) has excellent potency (1.3 nM), MCT1 selectivity (>1000×), secondary pharmacology, clean mechanism of action, suitable properties for oral administration in the clinic, and good preclinical efficacy in combination with cediranib.

The Global Characterisation of a Drug-Dendrimer Conjugate - PEGylated poly-lysine Dendrimer.

The recent emergence of drug-dendrimer conjugates within pharmaceutical industry research and development introduces a range of challenges for analytical and measurement science. These molecules are very high molecular weight (100-200kDa) with a significant degree of structural complexity. The characteristics and quality attributes that require understanding and definition, and impact efficacy and safety, are diverse. They relate to the intact conjugate, the various building blocks of these complex systems and the level of the free and bound active pharmaceutical ingredient (API). From an analytical and measurement science perspective, this necessitates the measurement of the molecular weight, impurity characterisation, the quantitation of the number of conjugated versus free API molecules, the determination of the impurity profiles of the building blocks, primary structure and both particle size and morphology. Here we report the first example of a global characterisation of a drug-dendrimer conjugate - PEGylated poly-lysine dendrimer currently under development (AZD0466). The impact of the wide variety of analytical and measurement techniques on the overall understanding of this complex molecular entity is discussed, with the relative capabilities of the various approaches compared. The results of this study are an essential platform for the research and development of the future generations of related dendrimer-based medicines.

Exploration of a Nitromethane-Carbonylation Strategy during Route Design of an Atropisomeric KRASG12C Inhibitor.

Route design and proof of concept synthesis was conducted on a synthetically challenging atropisomeric KRASG12C inhibitor to support clinical API manufacture. Improvements to the synthesis of a chiral piperazine fragment gave reduced step count and streamlined protecting group strategy via the formation and methanol ring opening of an N-carboxy-anhydride (NCA). The complex atropisomeric nitroquinoline was accessed via an early stage salt-resolution followed by a formal two-part nitromethane-carbonylation, avoiding a high temperature Gould-Jacobs cyclization that previously led to atropisomer racemization. The substrate scope of the formal nitromethane-carbonylation strategy was further explored for a range of ortho-substituted bromo/iodo unprotected anilines.

Machine learning meets mechanistic modelling for accurate prediction of experimental activation energies.

Accurate prediction of chemical reactions in solution is challenging for current state-of-the-art approaches based on transition state modelling with density functional theory. Models based on machine learning have emerged as a promising alternative to address these problems, but these models currently lack the precision to give crucial information on the magnitude of barrier heights, influence of solvents and catalysts and extent of regio- and chemoselectivity. Here, we construct hybrid models which combine the traditional transition state modelling and machine learning to accurately predict reaction barriers. We train a Gaussian Process Regression model to reproduce high-quality experimental kinetic data for the nucleophilic aromatic substitution reaction and use it to predict barriers with a mean absolute error of 0.77 kcal mol-1 for an external test set. The model was further validated on regio- and chemoselectivity prediction on patent reaction data and achieved a competitive top-1 accuracy of 86%, despite not being trained explicitly for this task. Importantly, the model gives error bars for its predictions that can be used for risk assessment by the end user. Hybrid models emerge as the preferred alternative for accurate reaction prediction in the very common low-data situation where only 100-150 rate constants are available for a reaction class. With recent advances in deep learning for quickly predicting barriers and transition state geometries from density functional theory, we envision that hybrid models will soon become a standard alternative to complement current machine learning approaches based on ground-state physical organic descriptors or structural information such as molecular graphs or fingerprints.

Tricyclic Indazoles-A Novel Class of Selective Estrogen Receptor Degrader Antagonists.

Herein, we report the identification and synthesis of a series of tricyclic indazoles as a novel class of selective estrogen receptor degrader antagonists. Replacement of a phenol, present in our previously reported tetrahydroisoquinoline scaffold, with an indazole group led to the removal of a reactive metabolite signal in an in vitro glutathione trapping assay. Further optimization, guided by X-ray crystal structures and NMR conformational work, varied the alkyl side chain and pendant aryl group and resulted in compounds with low turnover in human hepatocytes and enhanced chemical stability. Compound 9 was profiled as a representative of the series in terms of pharmacology and demonstrated the desired estrogen receptor α degrader-antagonist profile and demonstrated activity in a xenograft model of breast cancer.

Solubility prediction from first principles: a density of states approach.

Solubility is a fundamental property of widespread significance. Despite its importance, its efficient and accurate prediction from first principles remains a major challenge. Here we propose a novel method to predict the solubility of molecules using a density of states (DOS) approach from classical molecular simulation. The method offers a potential route to solubility prediction for large (including drug-like) molecules over a range of temperatures and pressures, all from a modest number of simulations. The method was employed to predict the solubility of sodium chloride in water at ambient conditions, yielding a value of 3.77(5) mol kg-1. This is in close agreement with other approaches based on molecular simulation, the consensus literature value being 3.71(25) mol kg-1. The predicted solubility is about half of the experimental value, the disparity being attributed to the known limitation of the Joung-Cheatham force field model employed for NaCl. The proposed method also accurately predicted the NaCl model's solubility over the temperature range 298-373 K directly from the density of states data used to predict the ambient solubility.

AZD9496: An Oral Estrogen Receptor Inhibitor That Blocks the Growth of ER-Positive and ESR1-Mutant Breast Tumors in Preclinical Models.

Fulvestrant is an estrogen receptor (ER) antagonist administered to breast cancer patients by monthly intramuscular injection. Given its present limitations of dosing and route of administration, a more flexible orally available compound has been sought to pursue the potential benefits of this drug in patients with advanced metastatic disease. Here we report the identification and characterization of AZD9496, a nonsteroidal small-molecule inhibitor of ERα, which is a potent and selective antagonist and downregulator of ERα in vitro and in vivo in ER-positive models of breast cancer. Significant tumor growth inhibition was observed as low as 0.5 mg/kg dose in the estrogen-dependent MCF-7 xenograft model, where this effect was accompanied by a dose-dependent decrease in PR protein levels, demonstrating potent antagonist activity. Combining AZD9496 with PI3K pathway and CDK4/6 inhibitors led to further growth-inhibitory effects compared with monotherapy alone. Tumor regressions were also seen in a long-term estrogen-deprived breast model, where significant downregulation of ERα protein was observed. AZD9496 bound and downregulated clinically relevant ESR1 mutants in vitro and inhibited tumor growth in an ESR1-mutant patient-derived xenograft model that included a D538G mutation. Collectively, the pharmacologic evidence showed that AZD9496 is an oral, nonsteroidal, selective estrogen receptor antagonist and downregulator in ER(+) breast cells that could provide meaningful benefit to ER(+) breast cancer patients. AZD9496 is currently being evaluated in a phase I clinical trial. Cancer Res; 76(11); 3307-18. ©2016 AACR.

Optimization of a Novel Binding Motif to (E)-3-(3,5-Difluoro-4-((1R,3R)-2-(2-fluoro-2-methylpropyl)-3-methyl-2,3,4,9-tetrahydro-1H-pyrido[3,4-b]indol-1-yl)phenyl)acrylic Acid (AZD9496), a Potent and Orally Bioavailable Selective Estrogen Receptor Downregulator and Antagonist.

The discovery of an orally bioavailable selective estrogen receptor downregulator (SERD) with equivalent potency and preclinical pharmacology to the intramuscular SERD fulvestrant is described. A directed screen identified the 1-aryl-2,3,4,9-tetrahydro-1H-pyrido[3,4-b]indole motif as a novel, druglike ER ligand. Aided by crystal structures of novel ligands bound to an ER construct, medicinal chemistry iterations led to (E)-3-(3,5-difluoro-4-((1R,3R)-2-(2-fluoro-2-methylpropyl)-3-methyl-2,3,4,9-tetrahydro-1H-pyrido[3,4-b]indol-1-yl)phenyl)acrylic acid (30b, AZD9496), a clinical candidate with high oral bioavailability across preclinical species that is currently being evaluated in phase I clinical trials for the treatment of advanced estrogen receptor (ER) positive breast cancer.

Discovery of 4-amino-N-[(1S)-1-(4-chlorophenyl)-3-hydroxypropyl]-1-(7H-pyrrolo[2,3-d]pyrimidin-4-yl)piperidine-4-carboxamide (AZD5363), an orally bioavailable, potent inhibitor of Akt kinases.

Wide-ranging exploration of analogues of an ATP-competitive pyrrolopyrimidine inhibitor of Akt led to the discovery of clinical candidate AZD5363, which showed increased potency, reduced hERG affinity, and higher selectivity against the closely related AGC kinase ROCK. This compound demonstrated good preclinical drug metabolism and pharmacokinetics (DMPK) properties and, after oral dosing, showed pharmacodynamic knockdown of phosphorylation of Akt and downstream biomarkers in vivo, and inhibition of tumor growth in a breast cancer xenograft model.

Sulfonyl-morpholino-pyrimidines: SAR and development of a novel class of selective mTOR kinase inhibitor.

High throughput screening to identify inhibitors of the mTOR kinase revealed sulfonyl-morpholino-pyrimidine 1 as an attractive start point. The compound displayed good physicochemical properties and selectivity over related kinases such as PI3Kα. Library preparation of related analogs allowed the establishment of additional SAR understanding and in particular the requirement for a key hydrogen bond donor motif at the 4-position of the phenyl ring in compounds such as indole 19. Isosteric replacement of the indole functionality led to the identification of urea compounds such as 32 that show good levels of mTOR inhibition in both enzyme and cellular assays.

Automated QSAR with a Hierarchy of Global and Local Models.

We present an automated QSAR procedure that is used in AstraZeneca's AutoQSAR system. The approach involves automatically selecting the most predictive models from pools of both global and local models. The effectiveness of this QSAR modelling strategy is demonstrated with a retrospective study that uses a diverse selection of 9 early stage AstraZeneca drug discovery projects and 3 physicochemical endpoints: LogD; solubility and human plasma protein binding. We show that the strategy makes a statistically significant improvement to the accuracy of predictions when compared to an updating global strategy, and that the systematic biases inherent in the global model predictions are almost completely removed. This improvement is attributed to the model selection aspect of the strategy.

A combined spectroscopic and crystallographic approach to probing drug-human serum albumin interactions.

The displacement of probes that bind selectively to subdomains IIA or IIIA on human serum albumin (HSA) by competing compounds has been followed using fluorescence spectroscopy, and has therefore been used to assign a primary binding site for these compounds in the presence and absence of fatty acids. The crystal structures have also been solved for three compounds: a matched pair of carboxylic acids whose binding strength to HSA unexpectedly decreased as the lipophilicity increased; and a highly bound sulphonamide that appeared not to displace the probes in the displacement assay. The crystallography results support the findings from the fluorescence displacement assay. The results indicate that drug binding to subdomain IB might also be important location for certain compounds.

Novel thienopyrimidine and thiazolopyrimidine kinase inhibitors with activity against Tie-2 in vitro and in vivo.

The SAR and improvement in potency against Tie2 of novel thienopyrimidine and thiazolopyrimidine kinase inhibitors are reported. The crystal structure of one of these compounds bound to the Tie-2 kinase domain is consistent with the SAR. These compounds have moderate potency in cellular assays of Tie-2 inhibition, good physical properties, DMPK, and show evidence of in vivo inhibition of Tie-2.

Discovery of imidazole vinyl pyrimidines as a novel class of kinase inhibitors which inhibit Tie-2 and are orally bioavailable.

Tie-2 is a receptor tyrosine kinase which is involved in angiogenesis and thereby growth of human tumours. The discovery and SAR of a novel class of imidazole-vinyl-pyrimidine kinase inhibitors, which inhibit Tie-2 in vitro is reported. Their synthesis was carried out by condensation of imidazole aldehydes with methyl pyrimidines. These compounds are lead-like, with low molecular weight, good physical properties and oral bioavailability.

Scaffold hopping using clique detection applied to reduced graphs.

Similarity-based methods for virtual screening are widely used. However, conventional searching using 2D chemical fingerprints or 2D graphs may retrieve only compounds which are structurally very similar to the original target molecule. Of particular current interest then is scaffold hopping, that is, the ability to identify molecules that belong to different chemical series but which could form the same interactions with a receptor. Reduced graphs provide summary representations of chemical structures and, therefore, offer the potential to retrieve compounds that are similar in terms of their gross features rather than at the atom-bond level. Using only a fingerprint representation of such graphs, we have previously shown that actives retrieved were more diverse than those found using Daylight fingerprints. Maximum common substructures give an intuitively reasonable view of the similarity between two molecules. However, their calculation using graph-matching techniques is too time-consuming for use in practical similarity searching in larger data sets. In this work, we exploit the low cardinality of the reduced graph in graph-based similarity searching. We reinterpret the reduced graph as a fully connected graph using the bond-distance information of the original graph. We describe searches, using both the maximum common induced subgraph and maximum common edge subgraph formulations, on the fully connected reduced graphs and compare the results with those obtained using both conventional chemical and reduced graph fingerprints. We show that graph matching using fully connected reduced graphs is an effective retrieval method and that the actives retrieved are likely to be topologically different from those retrieved using conventional 2D methods.