Sandra RB Allerheiligen
Leveraging Quantitative and Systems Pharmacology in Drug Development: Enabling Decisions
Both academic and pharmaceutical researchers have applied numerous technologies to optimize drug discovery and development. Over the last 15 years, model based drug development, which is often referred to as pharmacometrics (more pharmacokinetic/ pharmacodynamic based), has improved the overall probability of success of clinical trials and has impacted corporate and regulatory decisions. While tremendous progress has been made, we are still lagging in our ability to fully integrate a variety of quantitative approaches: systems biology, translational medicine, pharmacometrics and quantitative mechanistic approaches to describing disease. Our tools, knowledge, as well as the disciplines themselves have a tendency to remain hidden within silos. As a result, decisions still tend to be scientifically intuitive and empirical rather than scientifically intuitive based on quantitative rigor. Assimilation of knowledge across discovery and development through integrated models is represented in Quantitative and Systems Pharmacology and offers emerging opportunities to optimize learning and further enable quantitative decisions. These merging sciences offer the ability to much more mechanistically describe diseases and with greater insight into sources of variability, thereby, enhancing predictive capability and ultimately, quantitatively informing decisions.
Combining Networks of Genes, Small Molecules and Phenotypes to Understand and Predict Drug Action
The Pharmacogenomics Knowledgebase (PharmGKB, http://www.pharmgkb.org/) is devoted to cataloging all known associations between human genetic variation and drug response phenotypes. It currently contains human-curated associations between more than 3400 genetic variants in 2000 genes, affecting nearly 1000 drugs. It contains more than 5000 annotated literature citations and presents 70 pharmacokinetic and pharmacodynamic pathways. In addition to providing information to the pharmacogenomics research community, the PharmGKB is also useful as a data source for data mining and machine learning applications. In particular, we have shown that the data in PharmGKB is sufficiently rich to allow prediction of new gene-drug interactions based on existing ones. Our method uses a network of genetic interactions to infer new associations based on known interactions between a drug (and compounds similar to it) and a gene (and genes that interact with it). On cross-validation, our predictive performance is between 70% and 80%. We have also shown that we can extract reliable genetic networks from the published literature using natural language processing techniques. These methods have very high specificity (low false positives) with reasonable sensitivity (false negatives). Thus, we can extract rich gene-drug-phenotype interaction networks from text and the PharmGKB, and use them as starting points for experimental validation. They are particularly useful in the analysis of noisy high throughput data sets such as expression or genome-wide association studies. We conclude that a systems view of the cell is a more robust way to characterize the interactions of small molecules. We seek to combine experimental and computational approaches for understanding drug response in the context of genetic networks.
Discovering Master Regulators of Human Malignancies and Their Chemical
Modulators via an Integrated Systems Biology Approach.
The availability of accurate, genome-wide maps of molecular interactions within specific cellular contexts is starting to allow the elucidation of regulatory modules controlling both physiologic and pathologic processes. Further investigation of these modules in human malignancies has revealed master regulators, whose role is to integrate a complex spectrum of genetic alteration into a relatively small number of distinct molecular phenotypes. Indeed, these master regulators do not typically harbor (epi) genetic alterations and are altered only functionally. We suggest that genes playing such key roles in integrating a variety of distinct aberrant (epi) genetic pattens constitute optimal drug targets and biomarkers. We have combined analysis of master regulators of human malignancies with the systems biology based elucidation of the mechanism of action of small molecules. The resulting integrated approach is being tested to identify chemical modulators of tumor progression and drug-resistance, with specific application to Diffuse Large B Cell Lymphoma (DLBCL), Glioblastoma, and Glucocorticoid resistance in T cell Acute Lymphoblastic Leukemia (T-ALL).
Eli Lilly and Company
Application of Drug-Disease Models in the Development of Anti-Diabetic Agents
Diabetes is a complex chronic disease characterized by hyperglycemia resulting from defects in glucose and insulin homeostatic regulation. The availability of well-established, readily quantifiable biomarkers of the disease facilitates the model-based approaches to the development of pharmacologic agents of diverse mechanisms of action. Models that incorporate glucose-insulin feedback regulation and dynamic control mechanisms have gained increasing use in pharmaceutical research and development. Through simulations, these models are essential tools used for multiple purposes: 1) to aid the evaluation of biological targets and the interaction of various pathways in the translational stage of development; 2) to gain better understanding of the disease states in patient subpopulations and the associated sources of variability in therapeutic responses; 3) to optimize therapeutic regimens or clinical trial design; 4) to differentiate safety and efficacy profiles of a novel agent from drugs in the same class or the standards of care; 5) to evaluate combination therapies. Depending on the stage of drug development, models can be adapted to predict both acute responses and long term outcome. A brief overview of the types of drug-disease models and case examples of their applications in drug development will be presented.
Kathleen M. Giacomini
University of California, San Francisco
Transporter Systems in Drug Disposition and Response
Membrane transporters in the liver and kidney are key determinants of drug disposition, response and toxicity. For biliary or renal secretion of drugs, transporters located on basolateral and apical membranes work together to mediate transepithelial flux across polarized hepatic and renal epithelia. For drug metabolism, transporters play key roles in controlling the access of drugs to enzymes. Thus understanding transporter systems including the networks of proteins that work with these systems in the liver and kidney is important in predicting variation in clinical drug disposition and response. In this, presentation I will focus on genomic and expression analyses of transporters in human liver and kidney samples. Several co-expression networks consisting of drug metabolizing enzymes and transporters in the liver will be described. These networks identify the transporters, enzymes and regulatory proteins that work together in sensing and responding to xenobiotics including many clinically used drugs. Single nucleotide polymorphisms that contribute to variation in expression levels (eSNPs) of transporters will be presented. Collectively, these studies shed new light on the networks of genes including transporters that function in drug elimination and on the factors that contribute to variation in drug disposition and response.
Dana-Farber Cancer Institute/ Harvard University
Screening and Rational Design Approaches for Developing Selective and Non-Selective Protein Kinase Inhibitors
DFG-out loop conformation characteristic of an inactive kinase (Type II inhibitors). Here we describe our chemical and biological approaches towards developing new kinase inhibitors of wild-type and mutant kinases. Focus will be placed on allosteric Bcr-Abl inhibitors, irreversible EGFR kinase inhibitors, Mps1 and Erk5 inhibitors. Rational approaches towards developing multi-targeted kinase inhibitors will also be discussed.
RES Group Inc.
Mechanistic Modeling Approaches In Population PKPD
While empirical & statistical methods remain the cornerstone of population PKPD modeling, recent efforts have shifted more toward mechanistic-based modeling approaches in the hopes of uncovering additional insight from (often underutilized) clinical data. Mechanistic-based approaches are challenging in that they typically require more data and diverse expertise, but potentially yield more powerful clinical predictions and better insight into the underlying biology of drug and disease. Ideally, the application of these two modeling approaches can be complementary, in that the approach chosen should ultimately depend on the exact nature of the point in question. The goal of this presentation is to discuss the emerging mechanistic side of population PKPD modeling with particular emphasis on new technologies being developed, current challenges, and application examples from ongoing clinical trials.
Donald E. Mager
University at Buffalo, SUNY
Occupancy-Driven Cell Signaling Model of CD20 Agonists to Predict Combination Chemotherapy in Mice
A focused mathematical model was developed, based on an integration of temporal patterns of drug exposure, receptor occupancy, and signal transduction, to predict the effects of rituximab, a CD20 agonist, given in combination with either rhApo2L or fenretinide, on the time-course of tumor growth kinetics in murine non-Hodgkins lymphoma xenografts. The final model reflects major regulatory mechanisms including the target-mediated pharmacokinetics of rituximab, modulation of pro-apoptotic intracellular signaling induced by CD20 occupancy, and the relative efficacy of death receptor isoforms. System parameters were either fixed to reported values or estimated from published data of single-agent experiments. Model predicted profiles of Ramos cell xenografts after co-administration of rituximab with rhApo2L or fenretinide in mice are in agreement with data reported in the literature and capture the apparent synergy of such regimens in a mechanistic manner. Our model has the potential for optimizing preclinical studies and translational research focused on combination chemotherapy with CD20 agonists.
Timothy J. Mitchison
Harvard Medical School
Pharmacodynamics of Pacitaxel by Intravital Imaging in Mice
Many classic cytotoxic chemotherapy agents are cell-cycle specific in their actions, making it is unclear why they can work remarkably well in certain patients with epithelial cancers, where drug sensitivity is high despite the fact that cell cycle progression rate is low. Paclitaxel is one such agent - in cell culture at least, it only kills cancer cells that transit through mitosis, yet it can be highly effective in a sub-set of slow-growing epithelial cancers. I will discuss development of an approach to monitoring the responses of single tumor cells to pacitaxel in situ by intravital imaging of xenografted tumors, using a subcutaneous HT1080 model where cell cycle progression rate is relatively slow. Cells are labelled by transfection with fluorescent proteins whose localization is diagnostic of cell-cycle state, injected into mice and then imaged transdermally through a window chamber by confocal microscopy. We have demonstrated that it is possible to image single cells in a macroscope tumor with sufficient resolution that the fates of individual chromosomes can be discriminated. I will discuss very recent results showing the response of these tumors to pacitaxel involves both an immediate response associated with passage through the cell cycle as well as a delayed response that may responsible for the greatest fraction of cell killing. I will discuss the potential relevance of these xenograft pharmacodynamic data to the limited cell-resolution data available from human patients during paclitaxel treatment.
Applying Engineering Principles to the Development of Novel Cancer Therapies
Combining quantitative biology and computational modeling provides a powerful toolkit to design novel therapies in a context dependent manner. We will provide multiple examples where we translated the insights gained from modeling and simulation into practice by engineering and testing novel, antibody-based therapeutics in the context of the computer simulations. Through this iterative process between computational modeling and antibody engineering, we gain a deeper understanding of the drug’s mechanism of action which allows us to design therapeutics with a specific tumor type in mind. This context specific design can subsequently be translated into the clinic.
University of California, San Francisco
A Chemical Lens on Pharmacological Networks
Chemically similar drugs may be recognized by targets unrelated by sequence or structure. Conversely, the ligand sets for these targets are strikingly similar. It may thus be possible to relate many protein targets based on the chemical similarities among the ligands that bind to them; such similarities may often be orthogonal to those calculated bioinformatically. To investigate this, we have compared sets of ligands annotated for over 2000 targets to one another. A statistically significant similarity score for each pair of ligand sets can be calculated once a model of random similarity is developed, and the sets mapped together based on this similarity. Although no biological information is used in calculating these maps, biologically sensible clusters emerge. These relationships allow one to predict previously unknown off-targets for established drugs and reagents. Over 30 of these predictions have been tested in collaboration with Bryan Roth, with potencies on the “off-targets” ranging from 1.2 nM to 14 μM. This “Similarity Ensemble Approach” (SEA; http://sea.docking.org/) may be used comprehensively on all known drugs against all targets with precedented ligands. Applications of the method to interpret phenotypic screens will also be considered.
Harvard Medical School
Applying Systems Biology Concepts to Pharmacological Problems
I will briefly present a bottom-up perspective on the potential impact of recently developed measurement and modeling methods in pre-clinical pharmacology. Integration of computation and experiments promises to provide much-needed insight into the response of cells to drugs, the mechanistic basis of sensitivity and insensitivity, and the origins of acquired resistance. It is my hope that this conference will help to clarify how modeling cell-based experiments, animal studies, and clinical trial data will facilitate translation from bench to bedside, thereby providing new insight into existing therapies and promoting the development of new ones.
Merck Research Laboratories
Modeling of Aβ Dynamics in Animals and Humans
The amyloid hypothesis contends that build-up of Aβ and its associated plaques in brain tissue leads to development of Alzheimer’s disease. The γ-secretase inhibitor, MK-0752, can acutely and significantly lower CSF Aβ40 concentrations in humans. In this talk the dynamics of Aβ production in brain and its disposition will be discussion through examination of models developed from preclinical and clinical studies of MK-0752. Topics to be discussed include development of PK/PD models of CSF Aβ40 in humans and cross-species scaling of CNS Aβ response through modeling:
PK/PD Model of CSF Aβ40 in Humans
A substantial time-delay was noted between peak drug concentrations in CSF (3-4 hr) and peak A? reduction in CSF (~12 hr). The objectives of this modeling work were to develop a mechanistic model to account for this time delay, to characterize the inhibition by MK-0752, and to predict the effect with chronic drug administration. A model incorporating slow flow of CSF to the lumbar region best accounted for data obtained in humans with the g-secretase inhibitor, MK-0752. The IC50 of MK-0752 against g-secretase was estimated as 22.3 mM. Simulations of once-daily administration of MK-0752 predict that an average Ab40 reduction of 10-54% would be obtained at steady-state with doses from 100-300 mg.
Cross-Species Scaling of CNS Aβ Response through Modeling
The objectives of this work wereto develop PK/PD models of CNS Aβ40 response to MK-0752 in 5 different species (3 rodent, monkey, and human) and to develop methods for cross-species scaling of the PK/PD relationship.Because the same core equation describing response in brain tissue was used in all models, a cross-species scaling approach for the key IC50 parameter could be developed despite substantial differences in the time-course of response across the species. Correction for plasma protein binding was required for successful inter-species scaling of IC50.
University of California San Diego
Systems Pharmacology and Insulin Resistance
A systems medicine approach to insulin resistance in humans will be presented. Cellular and tissue defects associated with insulin resistance are coincident with transcriptional abnormalities and are improved after insulin sensitization with thiazolidinedione (TZD) PPAR ligands. Based on study of human subjects we have identified molecular and functional characteristics of insulin resistant subjects and distinctions between TZD treatment responder and non-responder subjects. Insulin resistant subjects exhibited alterations in skeletal muscle (e.g., glycolytic flux and intramuscular adipocytes) and adipose tissue (e.g., mitochondrial metabolism and inflammation) that improved relative to TZD-induced insulin sensitization. We will discuss uniquely characterized coordinated cellular and tissue functional pathways that are characteristic of insulin resistance, TZD-induced insulin sensitization, and potential TZD responsiveness. In order to address the relative efficacies of TZDs, we also characterized the insulin sensitivity and multi-tissue gene expression profiles of lean and insulin-resistant, obese Zucker rats untreated or treated with one of four PPAR ligands (pioglitazone, rosiglitazone, troglitazone and AG035029). Adipose tissue profiles revealed ligand-selective modulation of inflammatory and branched chain amino acid metabolic pathways which correlated with ligand-specific insulin sensitizing potency. We also examined lean and insulin resistant mice treated with the TZDs and identified pathways that are commonly regulated in human and mouse subjects. Our study towards deciphering responder-non-responder effects in Insulin resistance at the systemic level serves as a paradigm for pharmacological studies in human subjects.
Piet Van der Graaf
Pfizer Global Research and Development
Systems Pharmacology: bridging PKPD and Systems Biology in Drug Discovery
Preclinical or translational pharmacokinetic-pharmacodynamic (PKPD) modelling and simulation (M&S) is a relatively new area in drug discovery and until recently was mainly restricted to academic research. However, it is increasingly being recognized that successful implementation of PKPD reasoning in early drug discovery could have an even greater impact on the overall efficiency and success of pharmaceutical research as comparable investments in late-stage modeling and simulation. This is because arguably the most significant challenge facing the pharmaceutical industry is compound attrition resulting from the failure of preclinical efficacy and safety model data to translate into human proof of mechanism/concept studies. Therefore, it has been suggested that PKPD M&S can also play a significant role in early preclinical drug discovery and can provide a framework for translational research which links, in a quantitative manner, the interactions between a drug (or combination of drugs), pharmacological targets, physiological pathways and, ultimately, integrated disease systems.
Not surprisingly, with increased interest in its relevance to preclinical research, PKPD has evolved towards a more mechanistic approach, and (semi)-mechanistic PKPD models are now advocated not only by academic and industrial researchers, but also by regulators. However, the development of mechanism-based methods for cross-species scaling of PD parameters is still in its infancy, although some recent examples have suggested that allometric scaling may be applicable to predict not only pharmacokinetics but also pharmacodynamic responses in humans from data obtained in preclinical in vitro and in vivo models. A recent development in this area is based on the growing realization that innovation could be dramatically catalyzed by creating synergy at the interface between Systems Biology and PKPD, two disciplines which until now have largely existed in ‘parallel universes’ with a limited track record of impactful collaboration. This has led to the emergence of Systems Pharmacology. Broadly speaking, this is the quantitative analysis of the dynamic interactions between drug(s) and a biological system. In other words, Systems Pharmacology aims to understand the behaviour of the system as a whole, as opposed to the behaviour of its individual constituents; thus it has become the interface between PKPD and Systems Biology. It applies the concepts of Systems Engineering, Systems Biology and PKPD to the study of complex biological systems through iteration between computational and/or mathematical modelling and experimentation. Biomeasures (i.e. quantitative information about system properties such as target expression levels and turnover dynamics) are key enablers for success in Systems Pharmacology in a similar way as Biomarkers have been for PKPD and more investments are required in this area. In conclusion, M&S in its broadest sense will only achieve its full potential in tackling R&D attrition when the different components and disciplines operating at different stages of the R&D cycle are fully integrated into a ‘enhanced quantitative drug discovery and development’ (EQD3) paradigm.
Human Interactomes in Physiology and Disease
For over half a century it has been conjectured that macromolecules form complex networks of functionally interacting components, and that the molecular mechanisms underlying most biological processes correspond to particular steady states adopted by such cellular networks. However, until recently, systems-level theoretical conjectures remained largely unappreciated, mainly because of lack of supporting experimental data. To generate the information necessary to eventually address how complex cellular networks relate to biology, we initiated, at the scale of the whole proteome, an integrated approach for modeling protein-protein interaction or “interactome” networks. Our main questions are: How are interactome networks organized at the scale of the whole cell? How can we uncover local and global features underlying this organization, and how are interactome networks modified in human disease, such as cancer?
UT Southwestern Medical Center
Cellular Heterogeneity in Models of Cancer and Metabolism: Which Differences Make a Difference?
Phenotypic heterogeneity in cellular populations has been widely observed. However, interpreting possible roles of cellular heterogeneity in mammalian biology and disease is an exceedingly challenging problem due to the complexity of possible cellular phenotypes, the large dimension of potential perturbations, and the lack of methods for separating meaningful biological information from noise. To investigate cellular heterogeneity, our lab has developed an approach for summarizing the single-cell phenotypes of cellular populations into human- and machine-interpretable “profiles” that can be used to investigate changes due to physiological, pathological, or environmental perturbations. We will discuss recent progress in interpreting the role of heterogeneity in models of cancer and metabolism.
Yaning Wang, Ph.D.
Food and Drug Administration
Quantitative Drug-Disease-Trial Models: Vision 2020
Quantitative disease-drug-trial models allow learning from prior experience and summarize the knowledge in a ready to apply format. Employing these models to plan future development is proposed as a powerful solution to improve pharmaceutical R&D productivity. The disease and trial models are, to a large extent, independent of the product, but the drug model is not. The goals are to apply the disease and trial models to future development and regulatory decisions, and publicly share them. We propose working definitions of these models, describe the various subcomponents, provide examples, and discuss the challenges and potential solutions for developing such models. Building useful disease-drug-trial models is a challenging task and cannot be achieved by any single organization. It requires a consorted effort by industry, academic, and regulatory scientists. We also describe the strategic goals of the FDA Pharmacometrics group.
National Institute of General Medical Sciences | 45
Center Drive MSC 6200 | Bethesda, MD 20892-6200
U.S. Department of Health and Human Services | National Institutes of Health | USA.gov
|Version: 1.7.0||Privacy Statement|