The human hepatocyte TXG-MAPr: gene co-expression network modules to support mechanism-based risk assessment.

Mechanism-based risk assessment is urged to advance and fully permeate into current safety assessment practices, possibly at early phases of drug safety testing. Toxicogenomics is a promising source of mechanisms-revealing data, but interpretative analysis tools specific for the testing systems (e.g. hepatocytes) are lacking. In this study, we present the TXG-MAPr webtool (available at https://txg-mapr.eu/WGCNA_PHH/TGGATEs_PHH/ ), an R-Shiny-based implementation of weighted gene co-expression network analysis (WGCNA) obtained from the Primary Human Hepatocytes (PHH) TG-GATEs dataset. The 398 gene co-expression networks (modules) were annotated with functional information (pathway enrichment, transcription factor) to reveal their mechanistic interpretation. Several well-known stress response pathways were captured in the modules, were perturbed by specific stressors and showed preservation in rat systems (rat primary hepatocytes and rat in vivo liver), with the exception of DNA damage and oxidative stress responses. A subset of 87 well-annotated and preserved modules was used to evaluate mechanisms of toxicity of endoplasmic reticulum (ER) stress and oxidative stress inducers, including cyclosporine A, tunicamycin and acetaminophen. In addition, module responses can be calculated from external datasets obtained with different hepatocyte cells and platforms, including targeted RNA-seq data, therefore, imputing biological responses from a limited gene set. As another application, donors' sensitivity towards tunicamycin was investigated with the TXG-MAPr, identifying higher basal level of intrinsic immune response in donors with pre-existing liver pathology. In conclusion, we demonstrated that gene co-expression analysis coupled to an interactive visualization environment, the TXG-MAPr, is a promising approach to achieve mechanistic relevant, cross-species and cross-platform evaluation of toxicogenomic data.

A pan-cancer transcriptome analysis identifies replication fork and innate immunity genes as modifiers of response to the CHK1 inhibitor prexasertib.

The combined influence of oncogenic drivers, genomic instability, and/or DNA damage repair deficiencies increases replication stress in cancer. Cells with high replication stress rely on the upregulation of checkpoints like those governed by CHK1 for survival. Previous studies of the CHK1 inhibitor prexasertib demonstrated activity across multiple cancer types. Therefore, we sought to (1) identify markers of prexasertib sensitivity and (2) define the molecular mechanism(s) of intrinsic and acquired resistance using preclinical models representing multiple tumor types. Our findings indicate that while cyclin E dysregulation is a driving mechanism of prexasertib response, biomarkers associated with this aberration lack sufficient predictive power to render them clinically actionable for patient selection. Transcriptome analysis of a pan-cancer cell line panel and in vivo models revealed an association between expression of E2F target genes and prexasertib sensitivity and identified innate immunity genes associated with prexasertib resistance. Functional RNAi studies supported a causal role of replication fork components as modulators of prexasertib response. Mechanisms that protect cells from oncogene-induced replication stress may safeguard tumors from such stress induced by a CHK1 inhibitor, resulting in acquired drug resistance. Furthermore, resistance to prexasertib may be shaped by innate immunity.

Aurora A-Selective Inhibitor LY3295668 Leads to Dominant Mitotic Arrest, Apoptosis in Cancer Cells, and Shows Potent Preclinical Antitumor Efficacy.

Although Aurora A, B, and C kinases share high sequence similarity, especially within the kinase domain, they function distinctly in cell-cycle progression. Aurora A depletion primarily leads to mitotic spindle formation defects and consequently prometaphase arrest, whereas Aurora B/C inactivation primarily induces polyploidy from cytokinesis failure. Aurora B/C inactivation phenotypes are also epistatic to those of Aurora A, such that the concomitant inactivation of Aurora A and B, or all Aurora isoforms by nonisoform-selective Aurora inhibitors, demonstrates the Aurora B/C-dominant cytokinesis failure and polyploidy phenotypes. Several Aurora inhibitors are in clinical trials for T/B-cell lymphoma, multiple myeloma, leukemia, lung, and breast cancers. Here, we describe an Aurora A-selective inhibitor, LY3295668, which potently inhibits Aurora autophosphorylation and its kinase activity in vitro and in vivo, persistently arrests cancer cells in mitosis, and induces more profound apoptosis than Aurora B or Aurora A/B dual inhibitors without Aurora B inhibition-associated cytokinesis failure and aneuploidy. LY3295668 inhibits the growth of a broad panel of cancer cell lines, including small-cell lung and breast cancer cells. It demonstrates significant efficacy in small-cell lung cancer xenograft and patient-derived tumor preclinical models as a single agent and in combination with standard-of-care agents. LY3295668, as a highly Aurora A-selective inhibitor, may represent a preferred approach to the current pan-Aurora inhibitors as a cancer therapeutic agent.

A Novel Open Access Web Portal for Integrating Mechanistic and Toxicogenomic Study Results.

Applying toxicogenomics to improving the safety profile of drug candidates and crop protection molecules is most useful when it identifies relevant biological and mechanistic information that highlights risks and informs risk mitigation strategies. Pathway-based approaches, such as gene set enrichment analysis, integrate toxicogenomic data with known biological process and pathways. Network methods help define unknown biological processes and offer data reduction advantages. Integrating the 2 approaches would improve interpretation of toxicogenomic information. Barriers to the routine application of these methods in genome-wide transcriptomic studies include a need for "hands-on" computer programming experience, the selection of 1 or more analysis methods (eg pathway analysis methods), the sensitivity of results to algorithm parameters, and challenges in linking differential gene expression to variation in safety outcomes. To facilitate adoption and reproducibility of gene expression analysis in safety studies, we have developed Collaborative Toxicogeomics, an open-access integrated web portal using the Django web framework. The software, developed with the Python programming language, is modular, extensible and implements "best-practice" methods in computational biology. New study results are compared with over 4000 rodent liver experiments from Drug Matrix and open TG-GATEs. A unique feature of the software is the ability to integrate clinical chemistry and histopathology-derived outcomes with results from gene expression studies, leading to relevant mechanistic conclusions. We describe its application by analyzing the effects of several toxicants on liver gene expression and exemplify application to predicting toxicity study outcomes upon chronic treatment from expression changes in acute-duration studies.

Aurora A Kinase Inhibition Is Synthetic Lethal with Loss of the RB1 Tumor Suppressor Gene.

Loss-of-function mutations in the retinoblastoma gene RB1 are common in several treatment-refractory cancers such as small-cell lung cancer and triple-negative breast cancer. To identify drugs synthetic lethal with RB1 mutation (RB1 mut), we tested 36 cell-cycle inhibitors using a cancer cell panel profiling approach optimized to discern cytotoxic from cytostatic effects. Inhibitors of the Aurora kinases AURKA and AURKB showed the strongest RB1 association in this assay. LY3295668, an AURKA inhibitor with over 1,000-fold selectivity versus AURKB, is distinguished by minimal toxicity to bone marrow cells at concentrations active against RB1 mut cancer cells and leads to durable regression of RB1 mut tumor xenografts at exposures that are well tolerated in rodents. Genetic suppression screens identified enforcers of the spindle-assembly checkpoint (SAC) as essential for LY3295668 cytotoxicity in RB1-deficient cancers and suggest a model in which a primed SAC creates a unique dependency on AURKA for mitotic exit and survival. SIGNIFICANCE: The identification of a synthetic lethal interaction between RB1 and AURKA inhibition, and the discovery of a drug that can be dosed continuously to achieve uninterrupted inhibition of AURKA kinase activity without myelosuppression, suggest a new approach for the treatment of RB1-deficient malignancies, including patients progressing on CDK4/6 inhibitors.See related commentary by Dick and Li, p. 169.This article is highlighted in the In This Issue feature, p. 151.

Abemaciclib Is Active in Preclinical Models of Ewing Sarcoma via Multipronged Regulation of Cell Cycle, DNA Methylation, and Interferon Pathway Signaling.

PURPOSE: Ewing sarcoma (ES) is a rare and highly malignant cancer that occurs in the bone and surrounding tissue of children and adolescents. The EWS/ETS fusion transcription factor that drives ES pathobiology was previously demonstrated to modulate cyclin D1 expression. In this study, we evaluated abemaciclib, a small-molecule CDK4 and CDK6 (CDK4 and 6) inhibitor currently under clinical investigation in pediatric solid tumors, in preclinical models of ES. EXPERIMENTAL DESIGN: Using Western blot, high-content imaging, flow cytometry, ELISA, RNA sequencing, and CpG methylation assays, we characterized the in vitro response of ES cell lines to abemaciclib. We then evaluated abemaciclib in vivo in cell line-derived xenograft (CDX) and patient-derived xenograft (PDX) mouse models of ES as either a monotherapy or in combination with chemotherapy. RESULTS: Abemaciclib induced quiescence in ES cell lines via a G1 cell-cycle block, characterized by decreased proliferation and reduction of Ki-67 and FOXM1 expression and retinoblastoma protein (RB) phosphorylation. In addition, abemaciclib reduced DNMT1 expression and promoted an inflammatory immune response as measured by cytokine secretion, antigen presentation, and interferon pathway upregulation. Single-agent abemaciclib reduced ES tumor volume in preclinical mouse models and, when given in combination with doxorubicin or temozolomide plus irinotecan, durable disease control was observed. CONCLUSIONS: Collectively, our data demonstrate that the antitumor effects of abemaciclib in preclinical ES models are multifaceted and include cell-cycle inhibition, DNA demethylation, and immunogenic changes.

Genomic Aberrations that Activate D-type Cyclins Are Associated with Enhanced Sensitivity to the CDK4 and CDK6 Inhibitor Abemaciclib.

Most cancers preserve functional retinoblastoma (Rb) and may, therefore, respond to inhibition of D-cyclin-dependent Rb kinases, CDK4 and CDK6. To date, CDK4/6 inhibitors have shown promising clinical activity in breast cancer and lymphomas, but it is not clear which additional Rb-positive cancers might benefit from these agents. No systematic survey to compare relative sensitivities across tumor types and define molecular determinants of response has been described. We report a subset of cancers highly sensitive to CDK4/6 inhibition and characterized by various genomic aberrations known to elevate D-cyclin levels and describe a recurrent CCND1 3'UTR mutation associated with increased expression in endometrial cancer. The results suggest multiple additional classes of cancer that may benefit from CDK4/6-inhibiting drugs such as abemaciclib.

Mouse PDX Trial Suggests Synergy of Concurrent Inhibition of RAF and EGFR in Colorectal Cancer with BRAF or KRAS Mutations.

Purpose: To evaluate the antitumor efficacy of cetuximab in combination with LSN3074753, an analog of LY3009120 and pan-RAF inhibitor in 79 colorectal cancer patient-derived xenograft (PDX) models.Experimental Design: Seventy-nine well-characterized colorectal cancer PDX models were employed to conduct a single mouse per treatment group (n = 1) trial.Results: Consistent with clinical results, cetuximab was efficacious in wild-type KRAS and BRAF PDX models, with an overall response rate of 6.3% and disease control rate (DCR) of 20.3%. LSN3074753 was active in a small subset of PDX models that harbored KRAS or BRAF mutations. However, the combination treatment displayed the enhanced antitumor activity with DCR of 35.4%. Statistical analysis revealed that BRAF and KRAS mutations were the best predictors of the combinatorial activity and were significantly associated with synergistic effect with a P value of 0.01 compared with cetuximab alone. In 12 models with BRAF mutations, the combination therapy resulted in a DCR of 41.7%, whereas either monotherapy had a DCR of 8.3%. Among 44 KRAS mutation models, cetuximab or LSN3074753 monotherapy resulted in a DCR of 13.6% or 11.4%, respectively, and the combination therapy increased DCR to 34.1%. Molecular analysis suggests that EGFR activation is a potential feedback and resistant mechanism of pan-RAF inhibition.Conclusions: MAPK and EGFR pathway activations are two major molecular hallmarks of colorectal cancer. This mouse PDX trial recapitulated clinical results of cetuximab. Concurrent EGFR and RAF inhibition demonstrated synergistic antitumor activity for colorectal cancer PDX models with a KRAS or BRAF mutation. Clin Cancer Res; 23(18); 5547-60. ©2017 AACR.

Inhibition of RAF Isoforms and Active Dimers by LY3009120 Leads to Anti-tumor Activities in RAS or BRAF Mutant Cancers.

LY3009120 is a pan-RAF and RAF dimer inhibitor that inhibits all RAF isoforms and occupies both protomers in RAF dimers. Biochemical and cellular analyses revealed that LY3009120 inhibits ARAF, BRAF, and CRAF isoforms with similar affinity, while vemurafenib or dabrafenib have little or modest CRAF activity compared to their BRAF activities. LY3009120 induces BRAF-CRAF dimerization but inhibits the phosphorylation of downstream MEK and ERK, suggesting that it effectively inhibits the kinase activity of BRAF-CRAF heterodimers. Further analyses demonstrated that LY3009120 also inhibits various forms of RAF dimers including BRAF or CRAF homodimers. Due to these unique properties, LY3009120 demonstrates minimal paradoxical activation, inhibits MEK1/2 phosphorylation, and exhibits anti-tumor activities across multiple models carrying KRAS, NRAS, or BRAF mutation.

The role of translational bioinformatics in drug discovery.

The application of translational approaches (e.g. from bed to bench and back) is gaining momentum in the pharmaceutical industry. By utilizing the rapidly increasing volume of data at all phases of drug discovery, translational bioinformatics is poised to address some of the key challenges faced by the industry. Indeed, computational analysis of clinical data and patient records has informed decision-making in multiple aspects of drug discovery and development. Here, we review key examples of translational bioinformatics approaches to emphasize its potential to enhance the quality of drug discovery pipelines, reduce attrition rates and, ultimately, lead to more effective treatments.

Leveraging health social networking communities in translational research.

Health social networking communities are emerging resources for translational research. We have designed and implemented a framework called HyGen, which combines Semantic Web technologies, graph algorithms and user profiling to discover and prioritize novel associations across disciplines. This manuscript focuses on the key strategies developed to overcome the challenges in handling patient-generated content in Health social networking communities. Heuristic and quantitative evaluations were carried out in colorectal cancer. The results demonstrate the potential of our approach to bridge silos and to identify hidden links among clinical observations, drugs, genes and diseases. In Amyotrophic Lateral Sclerosis case studies, HyGen has identified 15 of the 20 published disease genes. Additionally, HyGen has highlighted new candidates for future investigations, as well as a scientifically meaningful connection between riluzole and alcohol abuse.

Structure-guided expansion of kinase fragment libraries driven by support vector machine models.

This work outlines a new de novo design process for the creation of novel kinase inhibitor libraries. It relies on a profiling paradigm that generates a substantial amount of kinase inhibitor data from which highly predictive QSAR models can be constructed. In addition, a broad diversity of X-ray structure information is needed for binding mode prediction. This is important for scaffold and substituent site selection. Borrowing from FBDD, the process involves fragmentation of known actives, proposition of binding mode hypotheses for the fragments, and model-driven recombination using a pharmacophore derived from known kinase inhibitor structures. The support vector machine method, using Merck atom pair derived fingerprint descriptors, was used to build models from activity from 6 kinase assays. These models were qualified prospectively by selecting and testing compounds from the internal compound collection. Overall hit and enrichment rates of 82% and 2.5%, respectively, qualified the models for use in library design. Using the process, 7 novel libraries were designed, synthesized and tested against these same 6 kinases. The results showed excellent results, yielding a 92% hit rate for the 179 compounds that made up the 7 libraries. The results of one library designed to include known literature compounds, as well as an analysis of overall substituent frequency, are discussed.

GPU accelerated support vector machines for mining high-throughput screening data.

Support Vector Machine (SVM), one of the most promising tools in chemical informatics, is time-consuming for mining large high-throughput screening (HTS) data sets. Here, we describe a parallelization of SVM-light algorithm on a graphic processor unit (GPU), using molecular fingerprints as descriptors and the Tanimoto index as kernel function. Comparison experiments based on six PubChem Bioassay data sets show that the GPU version is 43-104x faster than SVM-light for building classification models and 112-212x over SVM-light for building regression models.

Kinase inhibitor data modeling and de novo inhibitor design with fragment approaches.

A reconstructive approach based on computational fragmentation of existing inhibitors and validated kinase potency models to recombine and create "de novo" kinase inhibitor small molecule libraries is described. The screening results from model selected molecules from the corporate database and seven computationally derived small molecule libraries were used to evaluate this approach. Specifically, 1895 model selected database molecules were screened at 20 microM in six kinase assays and yielded an overall hit rate of 84%. These models were then used in the de novo design of seven chemical libraries consisting of 20-50 compounds each. Then 179 compounds from synthesized libraries were tested against these six kinases with an overall hit rate of 92%. Comparing predicted and observed selectivity profiles serves to highlight the strengths and limitations of the methodology, while analysis of functional group contributions from the libraries suggest general principles governing binding of ATP competitive compounds.