Paul B. Watkins

Case Definition and Phenotype Standardization in Drug-Induced Liver Injury

Drug‐induced liver injury (DILI) is the most frequent reason cited for the withdrawal of approved drugs from the market and accounts for up to 15% of the cases of acute liver failure. Investigators around the globe have begun to identify and study patients with DILI; several large registries and tissue banks are being established. In order to gain the maximum scientific benefit from these efforts, the definitions and terminology related to the clinical phenotypes of DILI must be harmonized. For this purpose, an international DILI Expert Working Group of clinicians and scientists reviewed current DILI terminology and diagnostic criteria so as to develop more uniform criteria that would define and characterize the spectrum of clinical syndromes that constitute DILI. Consensus was established with respect to the threshold criteria for definition of a case as being DILI, the pattern of liver injury, causality assessment, severity, and chronicity. Consensus was also reached on approaches to characterizing DILI in the setting of chronic liver diseases, including autoimmune hepatitis (AIH).

Susceptibility to Amoxicillin-Clavulanate-Induced Liver Injury is Influenced by Multiple HLA Class I and II Alleles

BACKGROUND & AIMS Drug-induced liver injury (DILI), especially from antimicrobial agents, is an important cause of serious liver disease. Amoxicillin-clavulanate (AC) is a leading cause of idiosyncratic DILI, but little is understood about genetic susceptibility to this adverse reaction. METHODS We performed a genome-wide association study using 822,927 single nucleotide polymorphism (SNP) markers from 201 White European and US cases of DILI following AC administration (AC-DILI) and 532 population controls, matched for genetic background. RESULTS AC-DILI was associated with many loci in the major histocompatibility complex. The strongest effect was with an HLA class II SNP (rs9274407, P=4.8×10(-14)), which correlated with rs3135388, a tag SNP of HLA-DRB1*1501-DQB1*0602 that was previously associated with AC-DILI. Conditioned on rs3135388, rs9274407 is still significant (P=1.1×10(-4)). An independent association was observed in the class I region (rs2523822, P=1.8×10(-10)), related to HLA-A*0201. The most significant class I and II SNPs showed statistical interaction (P=.0015). High-resolution HLA genotyping (177 cases and 219 controls) confirmed associations of HLA-A*0201 (P=2×10(-6)) and HLA-DQB1*0602 (P=5×10(-10)) and their interaction (P=.005). Additional, population-dependent effects were observed in HLA alleles with nominal significance. In an analysis of autoimmune-related genes, rs2476601 in the gene PTPN22 was associated (P=1.3×10(-4)). CONCLUSIONS Class I and II HLA genotypes affect susceptibility to AC-DILI, indicating the importance of the adaptive immune response in pathogenesis. The HLA genotypes identified will be useful in studies of the pathogenesis of AC-DILI but have limited utility as predictive or diagnostic biomarkers because of the low positive predictive values.

Quantitative analyses and transcriptomic profiling of circulating messenger RNAs as biomarkers of rat liver injury

Serum aminotransferases have been the clinical standard for evaluating liver injury for the past 50‐60 years. These tissue enzymes lack specificity, also tracking injury to other tissues. New technologies assessing tissue‐specific messenger RNA (mRNA) release into blood should provide greater specificity and permit indirect assessment of gene expression status of injured tissue. To evaluate the potential of circulating mRNAs as biomarkers of liver injury, rats were treated either with hepatotoxic doses of D‐(+)‐galactosamine (DGAL) or acetaminophen (APAP) or a myotoxic dose of bupivacaine HCl (BPVC). Plasma, serum, and liver samples were obtained from each rat. Serum alanine aminotransferase and aspartate aminotransferase were increased by all three compounds, whereas circulating liver‐specific mRNAs were only increased by the hepatotoxicants. With APAP, liver‐specific mRNAs were significantly increased in plasma at doses that had no effect on serum aminotransferases or liver histopathology. Characterization of the circulating mRNAs by sucrose density gradient centrifugation revealed that the liver‐specific mRNAs were associated with both necrotic debris and microvesicles. DGAL treatment also induced a shift in the size of plasma microvesicles, consistent with active release of microvesicles following liver injury. Finally, gene expression microarray analysis of the plasma following DGAL and APAP treatment revealed chemical‐specific profiles. Conclusion: The comparative analysis of circulating liver mRNAs with traditional serum transaminases and histopathology indicated that the circulating liver mRNAs were more specific and more sensitive biomarkers of liver injury. Further, the possibility of identifying chemical‐specific transcriptional profiles from circulating mRNAs could open a range of possibilities for identifying the etiology of drug/chemical‐induced liver injury. HEPATOLOGY 2010

FutureTox II: In vitro Data and In Silico Models for Predictive Toxicology

FutureTox II, a Society of Toxicology Contemporary Concepts in Toxicology workshop, was held in January, 2014. The meeting goals were to review and discuss the state of the science in toxicology in the context of implementing the NRC 21st century vision of predicting in vivo responses from in vitro and in silico data, and to define the goals for the future. Presentations and discussions were held on priority concerns such as predicting and modeling of metabolism, cell growth and differentiation, effects on sensitive subpopulations, and integrating data into risk assessment. Emerging trends in technologies such as stem cell-derived human cells, 3D organotypic culture models, mathematical modeling of cellular processes and morphogenesis, adverse outcome pathway development, and high-content imaging of in vivo systems were discussed. Although advances in moving towards an in vitro/in silico based risk assessment paradigm were apparent, knowledge gaps in these areas and limitations of technologies were identified. Specific recommendations were made for future directions and research needs in the areas of hepatotoxicity, cancer prediction, developmental toxicity, and regulatory toxicology.

Modeling Drug- and Chemical-Induced Hepatotoxicity with Systems Biology Approaches

We provide an overview of computational systems biology approaches as applied to the study of chemical- and drug-induced toxicity. The concept of “toxicity pathways” is described in the context of the 2007 US National Academies of Science report, “Toxicity testing in the 21st Century: A Vision and A Strategy.” Pathway mapping and modeling based on network biology concepts are a key component of the vision laid out in this report for a more biologically based analysis of dose-response behavior and the safety of chemicals and drugs. We focus on toxicity of the liver (hepatotoxicity) – a complex phenotypic response with contributions from a number of different cell types and biological processes. We describe three case studies of complementary multi-scale computational modeling approaches to understand perturbation of toxicity pathways in the human liver as a result of exposure to environmental contaminants and specific drugs. One approach involves development of a spatial, multicellular “virtual tissue” model of the liver lobule that combines molecular circuits in individual hepatocytes with cell–cell interactions and blood-mediated transport of toxicants through hepatic sinusoids, to enable quantitative, mechanistic prediction of hepatic dose-response for activation of the aryl hydrocarbon receptor toxicity pathway. Simultaneously, methods are being developing to extract quantitative maps of intracellular signaling and transcriptional regulatory networks perturbed by environmental contaminants, using a combination of gene expression and genome-wide protein-DNA interaction data. A predictive physiological model (DILIsym™) to understand drug-induced liver injury (DILI), the most common adverse event leading to termination of clinical development programs and regulatory actions on drugs, is also described. The model initially focuses on reactive metabolite-induced DILI in response to administration of acetaminophen, and spans multiple biological scales.

Exploring BSEP inhibition-mediated toxicity with a mechanistic model of drug-induced liver injury

Inhibition of the bile salt export pump (BSEP) has been linked to incidence of drug-induced liver injury (DILI), presumably by the accumulation of toxic bile acids in the liver. We have previously constructed and validated a model of bile acid disposition within DILIsym®, a mechanistic model of DILI. In this paper, we use DILIsym® to simulate the DILI response of the hepatotoxic BSEP inhibitors bosentan and CP-724,714 and the non-hepatotoxic BSEP inhibitor telmisartan in humans in order to explore whether we can predict that hepatotoxic BSEP inhibitors can cause bile acid accumulation to reach toxic levels. We also simulate bosentan in rats in order to illuminate potential reasons behind the lack of toxicity in rats compared to the toxicity observed in humans. DILIsym® predicts that bosentan, but not telmisartan, will cause mild hepatocellular ATP decline and serum ALT elevation in a simulated population of humans. The difference in hepatotoxic potential between bosentan and telmisartan is consistent with clinical observations. However, DILIsym® underpredicts the incidence of bosentan toxicity. DILIsym® also predicts that bosentan will not cause toxicity in a simulated population of rats, and that the difference between the response to bosentan in rats and in humans is primarily due to the less toxic bile acid pool in rats. Our simulations also suggest a potential synergistic role for bile acid accumulation and mitochondrial electron transport chain (ETC) inhibition in producing the observed toxicity in CP-724,714, and suggest that CP-724,714 metabolites may also play a role in the observed toxicity. Our work also compares the impact of competitive and noncompetitive BSEP inhibition for CP-724,714 and demonstrates that noncompetitive inhibition leads to much greater bile acid accumulation and potential toxicity. Our research demonstrates the potential for mechanistic modeling to contribute to the understanding of how bile acid transport inhibitors cause DILI.

Linking physiology to toxicity using DILIsym®, a mechanistic mathematical model of drug‐induced liver injury

The drug development industry faces multiple challenges in the realization of safe effective drugs. Computational modeling approaches can be used to support these efforts. One approach, mechanistic modeling, is new to the realm of drug safety. It holds the promise of not only predicting toxicity for novel compounds, but also illuminating the mechanistic underpinnings of toxicity. To increase the scientific communitys familiarity with mechanistic modeling in drug safety, this article seeks to provide perspective on the type of data used, how they are used and where they are lacking. Examples are derived from the development of DILIsym(®) software, a mechanistic model of drug-induced liver injury (DILI). DILIsym(®) simulates the mechanistic interactions and events from compound administration through the progression of liver injury and regeneration. Modeling mitochondrial toxicity illustrates the type and use of in vitro data to represent biological interactions, as well as insights on key differences between in vitro and in vivo conditions. Modeling bile acid toxicity illustrates a case in which the over-arching mechanism is well accepted, but many mechanistic details are lacking. Modeling was used to identify measurements predicted to strongly impact toxicity. Finally, modeling innate immune responses illustrates the importance of time-series data, particularly in the presence of positive and negative feedback loops, as well as the need for data from different animal species for better translation. These concepts are germane to most mechanistic models, although the details will vary. The use of mechanistic models is expected to improve the rational design of new drugs.

Mechanistic Modeling Reveals the Critical Knowledge Gaps in Bile Acid–Mediated DILI

Bile salt export pump (BSEP) inhibition has been proposed to be an important mechanism for drug‐induced liver injury (DILI). Modeling can prioritize knowledge gaps concerning bile acid (BA) homeostasis and thus help guide experimentation. A submodel of BA homeostasis in rats and humans was constructed within DILIsym, a mechanistic model of DILI. In vivo experiments in rats with glibenclamide were conducted, and data from these experiments were used to validate the model. The behavior of DILIsym was analyzed in the presence of a simulated theoretical BSEP inhibitor. BSEP inhibition in humans is predicted to increase liver concentrations of conjugated chenodeoxycholic acid (CDCA) and sulfate‐conjugated lithocholic acid (LCA) while the concentration of other liver BAs remains constant or decreases. On the basis of a sensitivity analysis, the most important unknowns are the level of BSEP expression, the amount of intestinal synthesis of LCA, and the magnitude of farnesoid‐X nuclear receptor (FXR)‐mediated regulation.

The new revolution in toxicology: the good, the bad, and the ugly.

In 2007, the United States National Academy of Sciences issued a report entitled Toxicity Testing in the 21st Century: A Vision and a Strategy. The report reviewed the state of the science and outlined a strategy for the future of toxicity testing. One of the more significant components of the vision established by the report was an emphasis on toxicity testing in human rather than animal systems. In the context of drug development, it is critical that the tools used to accomplish this strategy are maximally capable of evaluating human risk. Since 2007, many advances toward implementation of this vision have been achieved, particularly with regard to safety assessment of new chemical entities intended for pharmaceutical use.

A Genome-Wide Association Study Identifies Potential Susceptibility Loci for Hepatotoxicity Due to Various Drugs

Idiosyncratic drug-induced liver injury (DILI) is a leading cause of morbidity and mortality due to medication use yet is poorly studied, in part, due to its low incidence in the general population. Identification of genetic variants associated with these uncommon and difficult to diagnose adverse events could provide clues to underlying mechanisms. Hypothesis: Common genetic variants exist that contribute to DILI susceptibility from multiple drugs. Methods: DNA obtained from 783 individuals of European ancestry who experienced DILI attributed to >200 individual drugs were genotypedwith the Illumina 1Mor 1Mduo beadchip. Source of DNA samples included the following DILI registries: DILIN (401), DILIGEN (242), EUDRAGENE (89), and Malaga (51). Genome Wide Association (GWA) was performed using population controls (n=3001). Potential associations were tested for replication in 190 independent cases from DILIN. Results: GWA revealed a strong association within the MHC region (p 108), which did not replicate in validation cohorts composed of sufficient cases due to the same drug or class. Conclusion: Our replicated association between hepatocellular DILI and STAT4 supports the emerging role of the immune system in DILI susceptibility and suggests that common genetic variation may contribute to DILI susceptibility from multiple drugs. However, the generally negative GWA results suggest that there may be a preponderance of drug-specific genetic risk factors and/or rare genetic variation underlying DILI susceptibility. Therefore, we have begun whole exome and whole genome sequencing of DILI cases caused by the most frequent drugs in our cohorts in search of rarer, high-penetrance drug-specific risk factors.