A useful tool for drug interaction evaluation: the University of Washington Metabolism and Transport Drug Interaction Database

Effect of Carbamazepine on the Pharmacokinetics of Erdafitinib in Healthy Participants

Erdafitinib, a selective and potent oral pan-FGFR inhibitor, is metabolized mainly through CYP2C9 and CYP3A4 enzymes. This phase 1, open-label, single-sequence, drug-drug interaction study evaluated the pharmacokinetics, safety, and tolerability of a single oral dose of erdafitinib alone and when co-administered with steady state oral carbamazepine, a dual inducer of CYP3A4 and CYP2C9, in 13 healthy adult participants (NCT04330248). Compared with erdafitinib administration alone, carbamazepine co-administration decreased total and free maximum plasma concentrations of erdafitinib (Cmax) by 35% (95% CI 30%-39%) and 22% (95% CI 17%-27%), respectively. The areas under the concentration-time curve over the time interval from 0 to 168 hours, to the last quantifiable data point, and to time infinity (AUC168h, AUClast, AUCinf), were markedly decreased for both total erdafitinib (56%-62%) and free erdafitinib (48%-55%). The safety profile of erdafitinib was consistent with previous clinical studies in healthy participants, with no new safety concerns when administered with or without carbamazepine. Co-administration with carbamazepine may reduce the activity of erdafitinib due to reduced exposure. Concomitant use of strong CYP3A4 inducers with erdafitinib should be avoided.

Toward improved predictions of pharmacokinetics of transported drugs in hepatic impairment: Insights from the extended clearance model

Hepatic impairment (HI) moderately (<5-fold) affects the systemic exposure (i.e., area under the plasma concentration-time curve [AUC]) of drugs that are substrates of the hepatic sinusoidal organic anion transporting polypeptide (OATP) transporters and are excreted unchanged in the bile and/or urine. However, the effect of HI on their AUC is much greater (>10-fold) for drugs that are also substrates of cytochrome P450 (CYP) 3A enzymes. Using the extended clearance model, through simulations, we identified the ratio of sinusoidal efflux clearance (CL) over the sum of metabolic and biliary CLs as important in predicting the impact of HI on the AUC of dual OATP/CYP3A substrates. Because HI may reduce hepatic CYP3A-mediated CL to a greater extent than biliary efflux CL, the greater the contribution of the former versus the latter, the greater the impact of HI on drug AUC ratio (AUCRHI ). Using physiologically-based pharmacokinetic modeling and simulation, we predicted relatively well the AUCRHI of OATP substrates that are not significantly metabolized (pitavastatin, rosuvastatin, valsartan, and gadoxetic acid). However, there was a trend toward underprediction of the AUCRHI of the dual OATP/CYP3A4 substrates fimasartan and atorvastatin. These predictions improved when the sinusoidal efflux CL of these two drugs was increased in healthy volunteers (i.e., before incorporating the effect of HI), and by modifying the directionality of its modulation by HI (i.e., increase or decrease). To accurately predict the effect of HI on AUC of hepatobiliary cleared drugs it is important to accurately predict all hepatobiliary pathways, including sinusoidal efflux CL.

Drug-drug interaction prediction: databases, web servers and computational models

In clinical treatment, two or more drugs (i.e. drug combination) are simultaneously or successively used for therapy with the purpose of primarily enhancing the therapeutic efficacy or reducing drug side effects. However, inappropriate drug combination may not only fail to improve efficacy, but even lead to adverse reactions. Therefore, according to the basic principle of improving the efficacy and/or reducing adverse reactions, we should study drug-drug interactions (DDIs) comprehensively and thoroughly so as to reasonably use drug combination. In this review, we first introduced the basic conception and classification of DDIs. Further, some important publicly available databases and web servers about experimentally verified or predicted DDIs were briefly described. As an effective auxiliary tool, computational models for predicting DDIs can not only save the cost of biological experiments, but also provide relevant guidance for combination therapy to some extent. Therefore, we summarized three types of prediction models (including traditional machine learning-based models, deep learning-based models and score function-based models) proposed during recent years and discussed the advantages as well as limitations of them. Besides, we pointed out the problems that need to be solved in the future research of DDIs prediction and provided corresponding suggestions.

Pharmacogenetic Guidelines for Psychotropic Drugs: Optimizing Prescriptions in Clinical Practice

The modalities for prescribing a psychotropic (dose and choice of molecule) are currently unsatisfactory, which can lead to a lack of efficacy of the treatment associated with prolonged exposure of the patient to the symptoms of his or her illness and the side effects of the molecule. In order to improve the quality of treatment prescription, a part of the current biomedical research is dedicated to the development of pharmacogenetic tools for individualized prescription. In this guideline, we will present the genes of interest with level 1 clinical recommendations according to PharmGKB for the two major families of psychotropics: antipsychotics and antidepressants. For antipsychotics, there are CYP2D6 and CYP3A4, and for antidepressants, CYP2B6, CYP2D6, and CYP2C19. The study will focus on describing the role of each gene, presenting the variants that cause functional changes, and discussing the implications for prescriptions in clinical practice.

Assessment of Drug-Drug Interaction Risk Between Intravenous Fentanyl and the Glecaprevir/Pibrentasvir Combination Regimen in Hepatitis C Patients Using Physiologically Based Pharmacokinetic Modeling and Simulations

INTRODUCTION

An unsafe injection practice is one of the major contributors to new hepatitis C virus (HCV) infections; thus, people who inject drugs are a key population to prioritize to achieve HCV elimination. The introduction of highly effective and well-tolerated pangenotypic direct-acting antivirals, including glecaprevir/pibrentasvir (GLE/PIB), has revolutionized the HCV treatment landscape. Glecaprevir is a weak cytochrome P450 3A4 (CYP3A4) inhibitor, so there is the potential for drug-drug interactions (DDIs) with some opioids metabolized by CYP3A4, such as fentanyl. This study estimated the impact of GLE/PIB on the pharmacokinetics of intravenous fentanyl by building a physiologically based pharmacokinetic (PBPK) model.

METHODS

A PBPK model was developed for intravenous fentanyl by incorporating published information on fentanyl metabolism, distribution, and elimination in healthy individuals. Three clinical DDI studies were used to verify DDIs within the fentanyl PBPK model. This model was integrated with a previously developed GLE/PIB PBPK model. After model validation, DDI simulations were conducted by coadministering GLE 300 mg + PIB 120 mg with a single dose of intravenous fentanyl (0.5 µg/kg).

RESULTS

The predicted maximum plasma concentration ratio between GLE/PIB + fentanyl and fentanyl alone was 1.00, and the predicted area under the curve ratio was 1.04, suggesting an increase of only 4% in fentanyl exposure.

CONCLUSION

The administration of a therapeutic dose of GLE/PIB has very little effect on the pharmacokinetics of intravenous fentanyl. This negligible increase would not be expected to increase the risk of fentanyl overdose beyond the inherent risks related to the amount and purity of the fentanyl received during recreational use.

Epigenetic Mechanisms Contribute to Intraindividual Variations of Drug Metabolism Mediated by Cytochrome P450 Enzymes

Significant interindividual and intraindividual variations on cytochrome P450 (CYP)-mediated drug metabolism exist in the general population globally. Genetic polymorphisms are one of the major contribution factors for interindividual variations, but epigenetic mechanisms mainly contribute to intraindividual variations, including DNA methylation, histone modifications, microRNAs, and long non-coding RNAs. The current review provides analysis of advanced knowledge in the last decade on contributions of epigenetic mechanisms to intraindividual variations on CYP-mediated drug metabolism in several situations, including (1) ontogeny, the developmental changes of CYP expression in individuals from neonates to adults; (2) increased activities of CYP enzymes induced by drug treatment; (3) increased activities of CYP enzymes in adult ages induced by drug treatment at neonate ages; and (4) decreased activities of CYP enzymes in individuals with drug-induced liver injury (DILI). Furthermore, current challenges, knowledge gaps, and future perspective of the epigenetic mechanisms in development of CYP pharmacoepigenetics are discussed. In conclusion, epigenetic mechanisms have been proven to contribute to intraindividual variations of drug metabolism mediated by CYP enzymes in age development, drug induction, and DILI conditions. The knowledge has helped understanding how intraindividual variation are generated. Future studies are needed to develop CYP-based pharmacoepigenetics to guide clinical applications for precision medicine with improved therapeutic efficacy and reduced risk of adverse drug reactions and toxicity. SIGNIFICANCE STATEMENT: Understanding epigenetic mechanisms in contribution to intraindividual variations of CYP-mediated drug metabolism may help to develop CYP-based pharmacoepigenetics for precision medicine to improve therapeutic efficacy and reduce adverse drug reactions and toxicity for drugs metabolized by CYP enzymes.

Use of In Vivo Imaging and Physiologically-Based Kinetic Modelling to Predict Hepatic Transporter Mediated Drug-Drug Interactions in Rats

Gadoxetate, a magnetic resonance imaging (MRI) contrast agent, is a substrate of organic-anion-transporting polypeptide 1B1 and multidrug resistance-associated protein 2. Six drugs, with varying degrees of transporter inhibition, were used to assess gadoxetate dynamic contrast enhanced MRI biomarkers for transporter inhibition in rats. Prospective prediction of changes in gadoxetate systemic and liver AUC (AUCR), resulting from transporter modulation, were performed by physiologically-based pharmacokinetic (PBPK) modelling. A tracer-kinetic model was used to estimate rate constants for hepatic uptake (khe), and biliary excretion (kbh). The observed median fold-decreases in gadoxetate liver AUC were 3.8- and 1.5-fold for ciclosporin and rifampicin, respectively. Ketoconazole unexpectedly decreased systemic and liver gadoxetate AUCs; the remaining drugs investigated (asunaprevir, bosentan, and pioglitazone) caused marginal changes. Ciclosporin decreased gadoxetate khe and kbh by 3.78 and 0.09 mL/min/mL, while decreases for rifampicin were 7.20 and 0.07 mL/min/mL, respectively. The relative decrease in khe (e.g., 96% for ciclosporin) was similar to PBPK-predicted inhibition of uptake (97-98%). PBPK modelling correctly predicted changes in gadoxetate systemic AUCR, whereas underprediction of decreases in liver AUCs was evident. The current study illustrates the modelling framework and integration of liver imaging data, PBPK, and tracer-kinetic models for prospective quantification of hepatic transporter-mediated DDI in humans.

Functional Characterization of Six SLCO1B1 (OATP1B1) Variants Observed in Finnish Individuals with a Psychotic Disorder

Variants in the SLCO1B1 (solute carrier organic anion transporter family member 1B1) gene encoding the OATP1B1 (organic anion transporting polypeptide 1B1) protein are associated with altered transporter function that can predispose patients to adverse drug effects with statin treatment. We explored the effect of six rare SLCO1B1 single nucleotide variants (SNVs) occurring in Finnish individuals with a psychotic disorder on expression and functionality of the OATP1B1 protein. The SUPER-Finland study has performed exome sequencing on 9381 individuals with at least one psychotic episode during their lifetime. SLCO1B1 SNVs were annotated with PHRED-scaled combined annotation-dependent (CADD) scores and the Ensembl variant effect predictor. In vitro functionality studies were conducted for the SNVs with a PHRED-scaled CADD score of >10 and predicted to be missense. To estimate possible changes in transport activity caused by the variants, transport of 2',7'-dichlorofluorescein (DCF) in OATP1B1-expressing HEK293 cells was measured. According to the findings, additional tests with rosuvastatin and estrone sulfate were conducted. The amount of OATP1B1 in crude membrane fractions was quantified using a liquid chromatography tandem mass spectrometry-based quantitative targeted absolute proteomics analysis. Six rare missense variants of SLCO1B1 were identified in the study population, located in transmembrane helix 3: c.317T>C (p.106I>T), intracellular loop 2: c.629G>T (p.210G>V), c.633A>G (p.211I>M), c.639T>A (p.213N>L), transmembrane helix 6: 820A>G (p.274I>V), and the C-terminal end: 2005A>C (p.669N>H). Of these variants, SLCO1B1 c.629G>T (p.210G>V) resulted in the loss of in vitro function, abolishing the uptake of DCF, estrone sulfate, and rosuvastatin and reducing the membrane protein expression to 31% of reference OATP1B1. Of the six rare missense variants, SLCO1B1 c.629G>T (p.210G>V) causes a loss of function of OATP1B1 transport in vitro and severely decreases membrane protein abundance. Carriers of SLCO1B1 c.629G>T might be susceptible to altered pharmacokinetics of OATP1B1 substrate drugs and might have increased likelihood of adverse drug effects such as statin-associated musculoskeletal symptoms.

Preclinical development of an EP2 antagonist for post-seizure cognitive deficits

Cognitive comorbidities can substantially reduce quality of life in people with epilepsy. Inflammation is a component of all chronic diseases including epilepsy, as well as acute events like status epilepticus (SE). Neuroinflammation is the consequence of several broad signaling cascades including cyclooxygenase-2 (COX-2)-associated pathways. Activation of the EP2 receptor for prostaglandin E2 appears responsible for blood-brain barrier leakage and much of the inflammatory reaction, neuronal injury and cognitive deficit that follows seizure-provoked COX-2 induction in brain. Here we show that brief exposure of mice to TG11-77, a potent, selective, orally available and brain permeant EP2 antagonist, eliminates the profound cognitive deficit in Y-maze performance after SE and reduces delayed mortality and microgliosis, with a minimum effective i.p. dose (as free base) of 8.8 mg/kg. All in vitro studies required to submit an investigational new drug (IND) application for TG11-77 have been completed, and non-GLP dose range-finding toxicology in the rat identified no overt, organ or histopathology signs of toxicity after 7 days of oral administration at 1000 mg/kg/day. Plasma exposure in the rat was dose-linear between 15 and 1000 mg/kg dosing. TG11-77 thus appears poised to continue development towards the initial clinical test of the hypothesis that EP2 receptor modulation after SE can provide the first preventive treatment for one of the chief comorbidities of epilepsy.

Can Mechanistic Static Models for Drug-Drug Interactions Support Regulatory Filing for Study Waivers and Label Recommendations?

BACKGROUND AND OBJECTIVE

Mechanistic static and dynamic physiologically based pharmacokinetic models are used in clinical drug development to assess the risk of drug-drug interactions (DDIs). Currently, the use of mechanistic static models is restricted to screening DDI risk for an investigational drug, while dynamic physiologically based pharmacokinetic models are used for quantitative predictions of DDIs to support regulatory filing. As physiologically based pharmacokinetic model development by sponsors as well as a review of models by regulators require considerable resources, we explored the possibility of using mechanistic static models to support regulatory filing, using representative cases of successful physiologically based pharmacokinetic submissions to the US Food and Drug Administration under different classes of applications.

METHODS

Drug-drug interaction predictions with mechanistic static models were done for representative cases in the different classes of applications using the same data and modelling workflow as described in the Food and Drug Administration clinical pharmacology reviews. We investigated the hypothesis that the use of unbound average steady-state concentrations of modulators as driver concentrations in the mechanistic static models should lead to the same conclusions as those from physiologically based pharmacokinetic modelling for non-dynamic measures of DDI risk assessment such as the area under the plasma concentration-time curve ratio, provided the same input data are employed for the interacting drugs.

RESULTS

Drug-drug interaction predictions of area under the plasma concentration-time curve ratios using mechanistic static models were mostly comparable to those reported in the Food and Drug Administration reviews using physiologically based pharmacokinetic models for all representative cases in the different classes of applications.

CONCLUSIONS

The results reported in this study should encourage the use of models that best fit an intended purpose, limiting the use of physiologically based pharmacokinetic models to those applications that leverage its unique strengths, such as what-if scenario testing to understand the effect of dose staggering, evaluating the role of uptake and efflux transporters, extrapolating DDI effects from studied to unstudied populations, or assessing the impact of DDIs on the exposure of a victim drug with concurrent mechanisms. With this first step, we hope to trigger a scientific discussion on the value of a routine comparison of the two methods for regulatory submissions to potentially create a best practice that could help identify examples where the use of dynamic changes in modulator concentrations could make a difference to DDI risk assessment.

Evaluating the performance of machine-learning regression models for pharmacokinetic drug-drug interactions

Combination therapy or concomitant drug administration can be associated with pharmacokinetic drug-drug interactions, increasing the risk of adverse drug events and reduced drug efficacy. Thus far, machine-learning models have been developed that can classify drug-drug interactions. However, to enable quantification of the pharmacokinetic effects of a drug-drug interaction, regression-based machine learning should be explored. Therefore, this study investigated the use of regression-based machine learning to predict changes in drug exposure caused by pharmacokinetic drug-drug interactions. Fold changes in exposure relative to substrate drug monotherapy were collected from 120 clinical drug-drug interaction studies extracted from the Washington Drug Interaction Database and SimCYP compound library files. Drug characteristics (features) were collected such as structure, physicochemical properties, in vitro pharmacokinetic properties, cytochrome P450 metabolic activity, and population characteristics. Three different regression-based supervised machine-learning models were then applied to the prediction task: random forest, elastic net, and support vector regressor. Model performance was evaluated using fivefold cross-validation. Strongest performance was observed with support vector regression, with 78% of predictions within twofold of the observed exposure changes. The results show that changes in drug exposure can be predicted with reasonable accuracy using regression-based machine-learning models trained on data available early in drug discovery. This has potential applications in enabling earlier drug-drug interaction risk assessment for new drug candidates.

Macrolide Treatment Failure due to Drug–Drug Interactions: Real-World Evidence to Evaluate a Pharmacological Hypothesis

Macrolide antibiotics have received criticism concerning their use and risk of treatment failure. Nevertheless, they are an important class of antibiotics and are frequently used in clinical practice for treating a variety of infections. This study sought to utilize pharmacoepidemiology methods and pharmacology principles to estimate the risk of macrolide treatment failure and quantify the influence of their pharmacokinetics on the risk of treatment failure, using clinically reported drug–drug interaction data. Using a large, commercial claims database (2006–2015), inclusion and exclusion criteria were applied to create a cohort of patients who received a macrolide for three common acute infections. Furthermore, an additional analysis examining only bacterial pneumonia events treated with macrolides was conducted. These criteria were formulated specifically to ensure treatment failure would not be expected nor influenced by intrinsic or extrinsic factors. Treatment failure rates were 6% within the common acute infections and 8% in the bacterial pneumonia populations. Regression results indicated that macrolide AUC changes greater than 50% had a significant effect on treatment failure risk, particularly for azithromycin. In fact, our results show that decreased or increased exposure change can influence failure risk, by 35% or 12%, respectively, for the acute infection scenarios. The bacterial pneumonia results were less significant with respect to the regression analyses. This integration of pharmacoepidemiology and clinical pharmacology provides a framework for utilizing real-world data to provide insight into pharmacokinetic mechanisms and support future study development related to antibiotic treatments.

Determining the adjusted initial treatment dose of warfarin anticoagulant medicine using kernel-based support vector regression

BACKGROUND AND OBJECTIVE

A novel research field in bioinformatics is pharmacogenomics and the corresponding applications of artificial intelligence tools. Pharmacogenomics is the study of the relationship between genotype and responses to medical measures such as drug use. One of the most effective drugs is warfarin anticoagulant, but determining its initial treatment dose is challenging. Mistakes in the determination of the initial treatment dose can result directly in patient death.

METHODS

Some of the most successful techniques for estimating the initial treatment dose are kernel-based methods. However, all the available studies use pre-defined and constant kernels that might not necessarily address the problem's intended requirements. The present study seeks to define and present a new computational kernel extracted from a data set. This process aims to utilize all the data-related statistical features to generate a dose determination tool proportional to the data set with minimum error rate. The kernel-based version of the least square support vector regression estimator was defined. Through this method, a more appropriate approach was proposed for predicting the adjusted dose of warfarin.

RESULTS AND CONCLUSION

This paper benefits from the International Warfarin Pharmacogenomics Consortium (IWPC) Database. The results obtained in this study demonstrate that the support vector regression with the proposed new kernel can successfully estimate the ideal dosage of warfarin for approximately 68% of patients.

A Quantitative Approach to the Prediction of Drug-Drug Interactions Mediated by Cytochrome P450 2C8 Inhibition

BACKGROUND

Ohno and Colleagues proposed an approach for predicting drug-drug interactions (DDIs) mediated by cytochrome P450 (CYP) 3A4 based on the use of the ratio of the inhibited to non-inhibited area under the plasma concentration time curve (AUC) of substrates to estimate the fraction of the dose metabolized via CYP3A4 (contribution ratio, CR) and the in vivo inhibitory potency of a perpetrator (inhibition ratio, IR). This study evaluated the performance of this approach on DDIs mediated by CYP2C8 inhibitors.

RESEARCH DESIGN AND METHODS

Initial estimates of CR and IR of CYP2C8 substrates and inhibitors were calculated for 33 DDI in vivo studies. The approach was externally validated with 17 additional studies. Bayesian orthogonal regression was used to refine the estimates of the parameters. Assessment of prediction success was conducted by plotting observed versus predicted AUC ratios.

RESULTS

Final estimates of CRs and IRs were obtained for 19 CYP2C8 substrates and 23 inhibitors, respectively. The method demonstrated good predictive capacity, with only two values outside of the prespecified limits.

CONCLUSIONS

The approach may help to adapt dose regimens for CYP2C8 substrates when given in combination with CYP2C8 inhibitors and to map the potential DDIs of new molecular entities.

Overview of Current Herb-Drug Interaction Databases

An HERB-Drug Interaction (HDI) database is a structured data collection method for HDI information extracted from scattered literatures for quick retrieval. Our review summarized the ten currently available HDI databases, including those databases comprising HDI on the market. A detailed comparison on the scope of monographs, including the nature of content extracted from the original literature and user interfaces of these databases, was performed, and the number of references of fifty popular herbs in each HDI database was counted and presented in a heatmap to give users an intuitive understanding of the focuses of different HDI databases. Since it is well known that the development and maintenance of databases need continuous investment of capital and manpower, the sustainability of these databases was also reviewed and compared. Recently, artificial intelligence (AI) technologies, especially Natural Language Processing (NLP), have been applied to screen specific topics from massive articles and automatically identify the names of drugs and herbs in the literature. However, its application on the labor-intensive extraction and evaluation of HDI-related experimental conditions and results from literature remains limited due to the scarcity of these HDI data and the lack of well-established annotated datasets for these specific NLP recognition tasks. In view of the difficulties faced by current HDI databases and potential expansion of AI application in HDI database development, we propose a standardized format for data reporting and use of Concept Unique Identifier (CUI) for medical terms in the literature to accelerate the structured data collection. SIGNIFICANCE STATEMENT: The worldwide popularity of botanical and/or traditional medicine products has raised safety concerns due to potential HDI. However, the publicly available HDI databases are mostly outdated or incomplete. Through our review of the currently available HDI databases, a clear understanding of the key issues could be obtained and possible solutions to overcome the labour-intensive extraction as well as professional evaluation of information in HDI database development are proposed.

How to Integrate CYP2D6 Phenoconversion Into Clinical Pharmacogenetics: A Tutorial

CYP2D6 genotype is increasingly being integrated into practice to guide prescribing of certain medications. The CYP2D6 drug metabolizing enzyme is susceptible to inhibition by concomitant drugs, which can lead to a clinical phenotype that is different from the genotype-based phenotype, a process referred to as phenoconversion. Phenoconversion is highly prevalent but not widely integrated into practice because of either limited experience on how to integrate or lack of knowledge that it has occurred. We built a calculator tool to help clinicians integrate a standardized method of assessing CYP2D6 phenoconversion into practice. During tool-building, we identified several clinical factors that need to be considered when implementing CYP2D6 phenoconversion into clinical practice. This tutorial shares the steps that the University of Florida Health Precision Medicine Program took to build the calculator tool and identified clinical factors to consider when implementing CYP2D6 phenoconversion in clinical practice.

Physicochemical Properties, Biotransformation, and Transport Pathways of Established and Newly Approved Medications: A Systematic Review of the Top 200 Most Prescribed Drugs vs. the FDA-Approved Drugs Between 2005 and 2016

BACKGROUND

Enzyme-mediated biotransformation of pharmacological agents is a crucial step in xenobiotic detoxification and drug disposition. Herein, we investigated the metabolism and physicochemical properties of the top 200 most prescribed drugs (established) as well as drugs approved by the US Food and Drug Administration (FDA) between 2005 and 2016 (newly approved).

OBJECTIVE

Our objective was to capture the changing trends in the routes of administration, physicochemical properties, and prodrug medications, as well as the contributions of drug-metabolizing enzymes and transporters to drug clearance.

METHODS

The University of Washington Drug Interaction Database (DIDB^®) as well as other online resources (e.g., CenterWatch.com, Drugs.com, DrugBank.ca, and PubChem.ncbi.nlm.nih.gov) was used to collect and stratify the dataset required for exploring the above-mentioned trends.

RESULTS

Analyses revealed that ~ 90% of all drugs in the established and newly approved drug lists were administered systemically (oral or intravenous). Meanwhile, the portion of biologics (molecular weight > 1 kDa) was 15 times greater in the newly approved list than established drugs. Additionally, there was a 4.5-fold increase in the number of compounds with a high calculated partition coefficient (cLogP > 3) and a high total polar surface area (> 75 Å²) in the newly approved drug vs. the established category. Further, prodrugs in established or newly approved lists were found to be converted to active compounds via hydrolysis, demethylases, and kinases. The contribution of cytochrome P450 (CYP) 3A4, as the major biotransformation pathway, has increased from 40% in the established drug list to 64% in the newly approved drug list. Moreover, the role of CYP1A2, CYP2C19, and CYP2D6 were decreased as major metabolizing enzymes among the newly approved medications. Among non-CYP major metabolizers, the contribution of alcohol dehydrogenases/aldehyde dehydrogenases (ADH/ALDH) and sulfotransferases decreased in the newly approved drugs compared with the established list. Furthermore, the highest contribution among uptake and efflux transporters was found for Organic Anion Transporting Polypeptide 1B1 (OATP1B1) and P-glycoprotein (P-gp), respectively.

CONCLUSIONS

The higher portion of biologics in the newly approved drugs compared with the established list confirmed the growing demands for protein- and antibody-based therapies. Moreover, the larger number of hydrophilic drugs found in the newly approved list suggests that the probability of toxicity is likely to decrease. With regard to CYP-mediated major metabolism, CYP3A5 showed an increased involvement owing to the identification of unique probe substrates to differentiate CYP3As. Furthermore, the contribution of OATP1B1 and P-gp did not show a significant shift in the newly approved drugs as compared to the established list because of their broad substrate specificity.

A New Data Repository for Pharmacokinetic Natural Product-Drug Interactions: From Chemical Characterization to Clinical Studies

There are many gaps in scientific knowledge about the clinical significance of pharmacokinetic natural product–drug interactions (NPDIs) in which the natural product (NP) is the precipitant and a conventional drug is the object. The National Center for Complimentary and Integrative Health created the Center of Excellence for NPDI Research (NaPDI Center) (www.napdi.org) to provide leadership and guidance on the study of pharmacokinetic NPDIs. A key contribution of the Center is the first user-friendly online repository that stores and links pharmacokinetic NPDI data across chemical characterization, metabolomics analyses, and pharmacokinetic in vitro and clinical experiments (repo.napdi.org). The design is expected to help researchers more easily arrive at a complete understanding of pharmacokinetic NPDI research on a particular NP. The repository will also facilitate multidisciplinary collaborations, as the repository links all of the experimental data for a given NP across the study types. The current work describes the design of the repository, standard operating procedures used to enter data, and pharmacokinetic NPDI data that have been entered to date. To illustrate the usefulness of the NaPDI Center repository, more details on two high-priority NPs, cannabis and kratom, are provided as case studies.

Use of Phenotypically Poor Metabolizer Individual Donor Human Liver Microsomes To Identify Selective Substrates of UGT2B10

UDP-glucuronosyltransferase (UGT)1A4 and UGT2B10 are the human UGT isoforms most frequently involved in N-glucuronidation of drugs. UGT2B10 exhibits higher affinity than UGT1A4 for numerous substrates, making it potentially the more important enzyme for metabolism of these compounds in vivo. Clinically relevant UGT2B10 polymorphisms, including a null activity splice site mutation common in African populations, can lead to large exposure differences for UGT2B10 substrates that may limit their developability as marketed drugs. UGT phenotyping approaches using recombinantly expressed UGTs are limited by low enzyme activity and lack of validation of scaling to in vivo. In this study, we describe the use of an efficient experimental protocol for identification of UGT2B10-selective substrates (i.e., those with high fraction metabolized by UGT2B10), which exploits the activity difference between pooled human liver microsomes (HLM) and HLM from a phenotypically UGT2B10 poor metabolizer donor. Following characterization of the approach with eight known UGT2B10 substrates, we used ligand-based virtual screening and literature precedents to select 24 potential UGT2B10 substrates of 140 UGT-metabolized drugs for testing. Of these, dothiepin, cidoxepin, cyclobenzaprine, azatadine, cyproheptadine, bifonazole, and asenapine were indicated to be selective UGT2B10 substrates that have not previously been described. UGT phenotyping experiments and tests comparing conjugative and oxidative clearance were then used to confirm these findings. These approaches provide rapid and sensitive ways to evaluate whether a potential drug candidate cleared via glucuronidation will be sensitive to UGT2B10 polymorphisms in vivo. SIGNIFICANCE STATEMENT: The role of highly polymorphic UDP-glucuronosyltransferase (UGT)2B10 is likely to be underestimated currently for many compounds cleared via N-glucuronidation due to high test concentrations often used in vitro and low activity of UGT2B10 preparations. The methodology described in this study can be combined with the assessment of UGT versus oxidative in vitro metabolism to rapidly identify compounds likely to be sensitive to UGT2B10 polymorphism (high fraction metabolized by UGT2B10), enabling either chemical modification or polymorphism risk assessment before candidate selection.

The Cancer Drug Fraction of Metabolism Database

This study aims to create a database for quantifying the fraction of metabolism of cytochrome P450 isozymes for cancer drugs approved by the US Food and Drug Administration. A reproducible data collection protocol was developed to extract essential information, including both substrate-depletion and metabolite-formation data from publicly available in vitro selective cytochrome P450 enzyme inhibition studies. We estimated the fraction of metabolism from the curated data. To demonstrate the utility of this database, we conducted an in vitro drug interaction prediction for the 42 cancer drugs. In the drug-drug interaction prediction, we identified 31 drug pairs with at least one cancer drug in each pair that had predicted area under concentration ratios > 2. We further found clinical drug interaction pieces of evidence in the literature to support 20 of these 31 drug-drug interaction pairs.

In Silico Toxicology Data Resources to Support Read-Across and (Q)SAR

A plethora of databases exist online that can assist in in silico chemical or drug safety assessment. However, a systematic review and grouping of databases, based on purpose and information content, consolidated in a single source, has been lacking. To resolve this issue, this review provides a comprehensive listing of the key in silico data resources relevant to: chemical identity and properties, drug action, toxicology (including nano-material toxicity), exposure, omics, pathways, Absorption, Distribution, Metabolism and Elimination (ADME) properties, clinical trials, pharmacovigilance, patents-related databases, biological (genes, enzymes, proteins, other macromolecules etc.) databases, protein-protein interactions (PPIs), environmental exposure related, and finally databases relating to animal alternatives in support of 3Rs policies. More than nine hundred databases were identified and reviewed against criteria relating to accessibility, data coverage, interoperability or application programming interface (API), appropriate identifiers, types of in vitro, in vivo,-clinical or other data recorded and suitability for modelling, read-across, or similarity searching. This review also specifically addresses the need for solutions for mapping and integration of databases into a common platform for better translatability of preclinical data to clinical data.

Clinical implications of an analysis of pharmacokinetics of crizotinib coadministered with dexamethasone in patients with non-small cell lung cancer

PURPOSE

Dexamethasone is a systemic corticosteroid and a known cytochrome P450 (CYP)3A inducer. Crizotinib is a selective tyrosine kinase inhibitor of ALK, ROS1, and MET and a substrate of CYP3A. This post hoc analysis characterized the use of concomitant CYP3A inducers with crizotinib and estimated the effect of dexamethasone use on crizotinib pharmacokinetics at steady state.

METHODS

This analysis used data from four clinical studies (PROFILE 1001, 1005, 1007, and 1014) including 1690 patients with non-small cell lung cancer with ALK or ROS1 rearrangements treated with crizotinib at 250 mg twice daily. Frequency and reasons for use of concomitant CYP3A inducers, including dexamethasone, with crizotinib were characterized. Multiple steady-state trough concentrations (C_trough,ss) of crizotinib were measured for each patient. A linear mixed-effects model was used for within-patient comparison of crizotinib C_trough,ss between dosing of crizotinib alone and crizotinib coadministered with dexamethasone consecutively for ≥ 21 days.

RESULTS

Dexamethasone was the most commonly used CYP3A inducer (30.4%). A total of 15 patients had crizotinib C_trough,ss for both crizotinib dosing with and without dexamethasone. The adjusted geometric mean ratio of crizotinib C_trough,ss following coadministration with dexamethasone relative to crizotinib without dexamethasone, as a percentage, was 98.2% (90% confidence interval, 79.1-122.0%).

CONCLUSIONS

Crizotinib plasma exposure following coadministration with dexamethasone was similar to that when crizotinib was administered without dexamethasone, indicating dexamethasone has no effect on crizotinib exposure or efficacy. Other CYP3A inducers with similar potency would likewise have no clinically relevant effect on crizotinib exposure.

Effect of Rifampicin on the Plasma Concentrations of Bile Acid-O-Sulfates in Monkeys and Human Liver-Transplanted Chimeric Mice With or Without Bile Flow Diversion

The present study examined the significance of enterohepatic circulation and the effect of rifampicin [an inhibitor of organic anion-transporting polypeptide 1B (OATP1B)] on the plasma concentrations of bile acid-O-sulfates (glycochenodeoxycholate-O-sulfate, lithocholate-O-sulfate, glycolithocholate-O-sulfate, and taurolithocholate-O-sulfate) in monkeys and human liver-transplanted chimeric mice (PXB mouse). Rifampicin significantly increased the area under the curve of bile acid-O-sulfates in monkeys (13-69 times) and PXB mice (13-25 times) without bile flow diversion. Bile flow diversion reduced the concentration of plasma bile acid-O-sulfates under control conditions in monkeys and the concentration of plasma glycochenodeoxycholate-O-sulfate in PXB mice. It also diminished diurnal variation of plasma lithocholate-O-sulfate, glycolithocholate-O-sulfate, and taurolithocholate-O-sulfate in PXB mice under control conditions. Bile flow diversion did not affect the plasma concentration of bile acid-O-sulfates in monkeys and PXB mice treated with rifampicin. Plasma coproporphyrin I and III levels were constant in monkeys throughout the study, even with bile flow diversion. This study demonstrated that bile acid-O-sulfates are endogenous OATP1B biomarkers in monkeys and PXB mice. Enterohepatic circulation can affect the baseline levels of plasma bile acid-O-sulfates and modify the effect of OATP1B inhibition.

Adverse Event Circumstances and the Case of Drug Interactions

Adverse events are a common and for the most part unavoidable consequence of therapeutic intervention. Nevertheless, available tomes of such data now provide us with an invaluable opportunity to study the relationship between human phenotype and drug-induced protein perturbations within a patient system. Deciphering the molecular basis of such adverse responses is not only paramount to the development of safer drugs but also presents a unique opportunity to dissect disease systems in search of novel response biomarkers, drug targets, and efficacious combination therapies. Inspired by the potential applications of this approach, we first examined adverse event circumstances reported in FAERS and then performed a molecular level interrogation of cancer patient adverse events to investigate the prevalence of drug-drug interactions in the context of patient responses. We discuss avoidable and/or preventable cases and how molecular analytics can help optimize therapeutic use of co-medications. While up to one out of three adverse events in this dataset might be explicable by iatrogenic, patient, and product/device related factors, almost half of the patients in FAERS received multiple drugs and one in four may have experienced effects attributable to drug interactions.

From Endogenous Compounds as Biomarkers to Plasma-Derived Nanovesicles as Liquid Biopsy; Has the Golden Age of Translational Pharmacokinetics-Absorption, Distribution, Metabolism, Excretion-Drug-Drug Interaction Science Finally Arrived?

It is now established that a drug's pharmacokinetics (PK) absorption, distribution, metabolism, excretion (ADME) and drug-drug interaction (DDI) profile can be modulated by age, disease, and genotype. In order to facilitate subject phenotyping and clinical DDI assessment, therefore, various endogenous compounds (in plasma and urine) have been pursued as drug-metabolizing enzyme and transporter biomarkers. Compared with biomarkers, however, the topic of circulating extracellular vesicles as "liquid biopsy" has received little attention within the ADME community; most organs secrete nanovesicles (e.g., exosomes) into the blood that contain luminal "cargo" derived from the originating organ (proteins, messenger RNA, and microRNA). As such, ADME profiling of plasma exosomes could be leveraged to better define genotype-phenotype relationships and the study of ontogeny, disease, and complex DDIs. If methods to support the isolation of tissue-derived plasma exosomes are successfully developed and validated, it is envisioned that they will be used jointly with genotyping, biomarkers, and modeling tools to greatly progress translational PK-ADME-DDI science.

Drug-Drug Interaction Extraction via Recurrent Hybrid Convolutional Neural Networks with an Improved Focal Loss

Drug-drug interactions (DDIs) may bring huge health risks and dangerous effects to a patient’s body when taking two or more drugs at the same time or within a certain period of time. Therefore, the automatic extraction of unknown DDIs has great potential for the development of pharmaceutical agents and the safety of drug use. In this article, we propose a novel recurrent hybrid convolutional neural network (RHCNN) for DDI extraction from biomedical literature. In the embedding layer, the texts mentioning two entities are represented as a sequence of semantic embeddings and position embeddings. In particular, the complete semantic embedding is obtained by the information fusion between a word embedding and its contextual information which is learnt by recurrent structure. After that, the hybrid convolutional neural network is employed to learn the sentence-level features which consist of the local context features from consecutive words and the dependency features between separated words for DDI extraction. Lastly but most significantly, in order to make up for the defects of the traditional cross-entropy loss function when dealing with class imbalanced data, we apply an improved focal loss function to mitigate against this problem when using the DDIExtraction 2013 dataset. In our experiments, we achieve DDI automatic extraction with a micro F-score of 75.48% on the DDIExtraction 2013 dataset, outperforming the state-of-the-art approach by 2.49%.

Capturing the applicability of in vitro-in silico membrane transporter data in chemical risk assessment and biomedical research

Costs, scientific and ethical concerns related to animal tests for regulatory decision-making have stimulated the development of alternative methods. When applying alternative approaches, kinetics have been identified as a key element to consider. Membrane transporters affect the kinetic processes of absorption, distribution, metabolism and excretion (ADME) of various compounds, such as drugs or environmental chemicals. Therefore, pharmaceutical scientists have intensively studied transporters impacting drug efficacy and safety. Besides pharmacokinetics, transporters are considered as major determinant of toxicokinetics, potentially representing an essential piece of information in chemical risk assessment. To capture the applicability of transporter data for kinetic-based risk assessment in non-pharmaceutical sectors, the EU Reference Laboratory for Alternatives to Animal Testing (EURL ECVAM) created a survey with a view of identifying the improvements needed when using in vitro and in silico methods. Seventy-three participants, from different sectors and with various kinds of expertise, completed the survey. The results revealed that transporters are investigated mainly during drug development, but also for risk assessment purposes of food and feed contaminants, industrial chemicals, cosmetics, nanomaterials and in the context of environmental toxicology, by applying both in vitro and in silico tools. However, to rely only on alternative methods for chemical risk assessment, it is critical that the data generated by in vitro and in silico methods are scientific integer, reproducible and of high quality so that they are trusted by decision makers and used by industry. In line, the respondents identified various challenges related to the interpretation and use of transporter data from non-animal methods. Overall, it was determined that a combined mechanistically-anchored in vitro-in silico approach, validated against available human data, would gain confidence in using transporter data within an animal-free risk assessment paradigm. Finally, respondents involved primarily in fundamental research expressed lower confidence in non-animal studies to unravel complex transporter mechanisms.

Developing User Personas to Aid in the Design of a User-Centered Natural Product-Drug Interaction Information Resource for Researchers

Pharmacokinetic interactions between natural products and conventional drugs can adversely impact patient outcomes. These complex interactions present unique challenges that require clear communication to researchers. We are creating a public information portal to facilitate researchers' access to credible evidence about these interactions. As part of a user-centered design process, three types of intended researchers were surveyed: drug-drug interaction scientists, clinical pharmacists, and drug compendium editors. Of the 23 invited researchers, 17 completed the survey. The researchers suggested a number of specific requirements for a natural product-drug interaction information resource, including specific information about a given interaction, the potential to cause adverse effects, and the clinical importance. Results were used to develop user personas that provided the development team with a concise and memorable way to represent information needs of the three main researcher types and a common basis for communicating the design's rationale.

Pharmacokinetic Interactions of Rolapitant With Cytochrome P450 3A Substrates in Healthy Subjects

Rolapitant (Varubi) is a neurokinin-1 receptor antagonist approved for the prevention of chemotherapy-induced nausea and vomiting. Rolapitant is primarily metabolized by the cytochrome P450 3A4 (CYP3A4) enzyme. Unlike other neurokinin-1 receptor antagonists, rolapitant is neither an inhibitor nor an inducer of CYP3A4 in vitro. The objective of this analysis was to examine the pharmacokinetics of rolapitant in healthy subjects and assess drug-drug interactions between rolapitant and midazolam (a CYP3A substrate), ketoconazole (a CYP3A inhibitor), or rifampin (a CYP3A4 inducer). Three phase 1, open-label, drug-drug interaction studies were conducted to examine the pharmacokinetic interactions of orally administered rolapitant with midazolam, rolapitant with ketoconazole, and rolapitant with rifampin. The pharmacokinetic profiles of midazolam and 1-hydroxy midazolam metabolites were essentially unchanged when coadministered with rolapitant, indicating the lack of a clinically relevant inhibition or induction of CYP3A by rolapitant. Coadministration of ketoconazole with rolapitant had no effects on rolapitant maximum concentration and resulted in an approximately 20% increase in the area under the concentration-time curve of rolapitant, suggesting that strong CYP3A inhibitors have minimal inhibitory effects on rolapitant exposure. Repeated administrations of rifampin appeared to reduce rolapitant exposure, resulting in a 33% decrease in maximum concentration and 87% decrease in area under the concentration-time curve from time zero to infinity. Coadministration of rolapitant did not affect the exposure of midazolam. Rifampin coadministration resulted in lower concentrations of rolapitant, and ketoconazole coadministration had no or minimal effects on rolapitant exposure. Rolapitant was safe and well tolerated when coadministered with ketoconazole, rifampin, or midazolam. No new safety signals were reported compared with previous studies of rolapitant.

Clinical Utility of Definitive Drug⁻Drug Interaction Testing in Primary Care

Drug–drug interactions (DDIs) are a leading cause of morbidity and mortality. New tools are needed to improve identification and treatment of DDIs. We conducted a randomized controlled trial to assess the clinical utility of a new test to identify DDIs and improve their management. Primary care physicians (PCPs) cared for simulated patients presenting with DDI symptoms from commonly prescribed medications and other ingestants. All physicians, in either control or one of two intervention groups, cared for six patients over two rounds of assessment. Intervention physicians were educated on the DDI test and given access to these test reports when caring for their patients in the second round. At baseline, we saw no significant differences in making the DDI diagnosis (p = 0.071) or DDI-related treatment (p = 0.640) between control and intervention arms. By round two, providers who accessed the DDI test performed significantly better in making the DDI diagnosis (+41.6%) and performing DDI-specific treatment (+12.2%) than in the previous round, and were 9.8 and 20.4 times more likely to diagnose and identify the DDI (p < 0.001 for all). The introduction of a definitive DDI test significantly increased identification, appropriate management, and counseling of DDIs among PCPs, which has the potential to improve clinical care.

Pharmacokinetics and Drug Interaction of Antiepileptic Drugs in Children and Adolescents

Selecting the most appropriate antiepileptic drug (AED) or combination of drugs for each patient and identifying the most suitable therapeutic regimen for their needs is increasingly challenging, especially among pediatric populations. In fact, the pharmacokinetics of several drugs vary widely in children with epilepsy because of age-related factors, which can influence the absorption, distribution, metabolism, and elimination of the pharmacological agent. In addition, individual factors, such as seizure type, associated comorbidities, individual pharmacokinetics, and potential drug interactions, may contribute to large fluctuations in serum drug concentrations and, therefore, clinical response. Therapeutic drug concentration monitoring (TDM) is an essential tool to deal with this complexity, enabling the definition of individual therapeutic concentrations and adaptive control of dosing to minimize drug interactions and prevent loss of efficacy or toxicity. Moreover, pharmacokinetic/pharmacodynamic modelling integrated with dashboard systems have recently been tested in antiepileptic therapy, although more clinical trials are required to support their use in clinical practice. We review the mechanism of action, pharmacokinetics, drug-drug interactions, and safety/tolerability profiles of the main AEDs currently used in children and adolescents, paying particular regard to issues of relevance when treating this patient population. Indications for TDM are provided for each AED as useful support to the clinical management of pediatric patients with epilepsy by optimizing pharmacological therapy.

Comparative Study of the Dose-Dependence of OATP1B Inhibition by Rifampicin Using Probe Drugs and Endogenous Substrates in Healthy Volunteers

PURPOSE

To evaluate association of the dose-dependent effect of rifampicin, an OATP1B inhibitor, on the plasma concentration-time profiles among OATP1B substrates drugs and endogenous substrates.

METHODS

Eight healthy volunteers received atorvastatin (1 mg), pitavastatin (0.2 mg), rosuvastatin (0.5 mg), and fluvastatin (2 mg) alone or with rifampicin (300 or 600 mg) in a crossover fashion. The plasma concentrations of these OATP1B probe drugs, total and direct bilirubin, glycochenodeoxycholate-3-sulfate (GCDCA-S), and coproporphyrin I, were determined.

RESULTS

The most striking effect of 600 mg rifampicin was on atorvastatin (6.0-times increase) and GCDCA-S (10-times increase). The AUC_0-24h of atorvastatin was reasonably correlated with that of pitavastatin (r² = 0.73) and with the AUC_0-4h of fluvastatin (r² = 0.62) and sufficiently with the AUC_0-24h of rosuvastatin (r² = 0.32). The AUC_0-24h of GCDCA-S was reasonably correlated with those of direct bilirubin (r² = 0.74) and coproporphyrin I (r² = 0.78), and sufficiently with that of total bilirubin (r² = 0.30). The AUC_0-24h of GCDCA-S, direct bilirubin, and coproporphyrin I were reasonably correlated with that of atorvastatin (r² = 0.48-0.70) [corrected].

CONCLUSION

These results suggest that direct bilirubin, GCDCA-S, and coproporphyrin I are promising surrogate probes for the quantitative assessment of potential OATP1B-mediated DDI.

Drug-drug interaction extraction via hierarchical RNNs on sequence and shortest dependency paths

Motivation

Adverse events resulting from drug-drug interactions (DDI) pose a serious health issue. The ability to automatically extract DDIs described in the biomedical literature could further efforts for ongoing pharmacovigilance. Most of neural networks-based methods typically focus on sentence sequence to identify these DDIs, however the shortest dependency path (SDP) between the two entities contains valuable syntactic and semantic information. Effectively exploiting such information may improve DDI extraction.

Results

In this article, we present a hierarchical recurrent neural networks (RNNs)-based method to integrate the SDP and sentence sequence for DDI extraction task. Firstly, the sentence sequence is divided into three subsequences. Then, the bottom RNNs model is employed to learn the feature representation of the subsequences and SDP, and the top RNNs model is employed to learn the feature representation of both sentence sequence and SDP. Furthermore, we introduce the embedding attention mechanism to identify and enhance keywords for the DDI extraction task. We evaluate our approach using the DDI extraction 2013 corpus. Our method is competitive or superior in performance as compared with other state-of-the-art methods. Experimental results show that the sentence sequence and SDP are complementary to each other. Integrating the sentence sequence with SDP can effectively improve the DDI extraction performance.

Availability and implementation

The experimental data is available at https://github.com/zhangyijia1979/hierarchical-RNNs-model-for-DDI-extraction.

Contact

zhyj@dlut.edu.cn or michel.dumontier@maastrichtuniversity.nl.

Supplementary information

Supplementary data are available at Bioinformatics online.

Progress in Prediction and Interpretation of Clinically Relevant Metabolic Drug-Drug Interactions: a Minireview Illustrating Recent Developments and Current Opportunities

PURPOSE OF REVIEW

This review gives a perspective on the current "state of the art" in metabolic drug-drug interaction (DDI) prediction. We highlight areas of successful prediction and illustrate progress in areas where limits in scientific knowledge or technologies prevent us from having full confidence.

RECENT FINDINGS

Several examples of success are highlighted. Work done for bitopertin shows how in vitro and clinical data can be integrated to give a model-based understanding of pharmacokinetics and drug interactions. The use of interpolative predictions to derive explicit dosage recommendations for untested DDIs is discussed using the example of ibrutinib, and the use of DDI predictions in lieu of clinical studies in new drug application packages is exemplified with eliglustat and alectinib. Alectinib is also an interesting case where dose adjustment is unnecessary as the activity of a major metabolite compensates sufficiently for changes in parent drug exposure. Examples where "unusual" cytochrome P450 (CYP) and non-CYP enzymes are responsible for metabolic clearance have shown the importance of continuing to develop our repertoire of in vitro regents and techniques. The time-dependent inhibition assay using human hepatocytes suspended in full plasma allowed improved DDI predictions, illustrating the importance of continued in vitro assay development and refinement.

SUMMARY

During the past 10 years, a highly mechanistic understanding has been developed in the area of CYP-mediated metabolic DDIs enabling the prediction of clinical outcome based on preclinical studies. The combination of good quality in vitro data and physiologically based pharmacokinetic modeling may now be used to evaluate DDI risk prospectively and are increasingly accepted in lieu of dedicated clinical studies.

Leveraging syntactic and semantic graph kernels to extract pharmacokinetic drug drug interactions from biomedical literature

Background

Information about drug–drug interactions (DDIs) supported by scientific evidence is crucial for establishing computational knowledge bases for applications like pharmacovigilance. Since new reports of DDIs are rapidly accumulating in the scientific literature, text-mining techniques for automatic DDI extraction are critical. We propose a novel approach for automated pharmacokinetic (PK) DDI detection that incorporates syntactic and semantic information into graph kernels, to address the problem of sparseness associated with syntactic-structural approaches. First, we used a novel all-path graph kernel using shallow semantic representation of sentences. Next, we statistically integrated fine-granular semantic classes into the dependency and shallow semantic graphs.

Results

When evaluated on the PK DDI corpus, our approach significantly outperformed the original all-path graph kernel that is based on dependency structure. Our system that combined dependency graph kernel with semantic classes achieved the best F-scores of 81.94 % for in vivo PK DDIs and 69.34 % for in vitro PK DDIs, respectively. Further, combining shallow semantic graph kernel with semantic classes achieved the highest precisions of 84.88 % for in vivo PK DDIs and 74.83 % for in vitro PK DDIs, respectively.

Conclusions

We presented a graph kernel based approach to combine syntactic and semantic information for extracting pharmacokinetic DDIs from Biomedical Literature. Experimental results showed that our proposed approach could extract PK DDIs from literature effectively, which significantly enhanced the performance of the original all-path graph kernel based on dependency structure.

Methods and strategies for assessing uncontrolled drug-drug interactions in population pharmacokinetic analyses: results from the International Society of Pharmacometrics (ISOP) Working Group

The purpose of this work was to present a consolidated set of guidelines for the analysis of uncontrolled concomitant medications (ConMed) as a covariate and potential perpetrator in population pharmacokinetic (PopPK) analyses. This white paper is the result of an industry-academia-regulatory collaboration. It is the recommendation of the working group that greater focus be given to the analysis of uncontrolled ConMeds as part of a PopPK analysis of Phase 2/3 data to ensure that the resulting outcome in the PopPK analysis can be viewed as reliable. Other recommendations include: (1) collection of start and stop date and clock time, as well as dose and frequency, in Case Report Forms regarding ConMed administration schedule; (2) prespecification of goals and the methods of analysis, (3) consideration of alternate models, other than the binary covariate model, that might more fully characterize the interaction between perpetrator and victim drug, (4) analysts should consider whether the sample size, not the percent of subjects taking a ConMed, is sufficient to detect a ConMed effect if one is present and to consider the correlation with other covariates when the analysis is conducted, (5) grouping of ConMeds should be based on mechanism (e.g., PGP-inhibitor) and not drug class (e.g., beta-blocker), and (6) when reporting the results in a publication, all details related to the ConMed analysis should be presented allowing the reader to understand the methods and be able to appropriately interpret the results.

Predicting drug metabolism: experiment and/or computation?

Drug metabolism can produce metabolites with physicochemical and pharmacological properties that differ substantially from those of the parent drug, and consequently has important implications for both drug safety and efficacy. To reduce the risk of costly clinical-stage attrition due to the metabolic characteristics of drug candidates, there is a need for efficient and reliable ways to predict drug metabolism in vitro, in silico and in vivo. In this Perspective, we provide an overview of the state of the art of experimental and computational approaches for investigating drug metabolism. We highlight the scope and limitations of these methods, and indicate strategies to harvest the synergies that result from combining measurement and prediction of drug metabolism.

Evolutionary Diagnosis of non-synonymous variants involved in differential drug response

Background

Many pharmaceutical drugs are known to be ineffective or have negative side effects in a substantial proportion of patients. Genomic advances are revealing that some non-synonymous single nucleotide variants (nsSNVs) may cause differences in drug efficacy and side effects. Therefore, it is desirable to evaluate nsSNVs of interest in their ability to modulate the drug response.

Results

We found that the available data on the link between drug response and nsSNV is rather modest. There were only 31 distinct drug response-altering (DR-altering) and 43 distinct drug response-neutral (DR-neutral) nsSNVs in the whole Pharmacogenomics Knowledge Base (PharmGKB). However, even with this modest dataset, it was clear that existing bioinformatics tools have difficulties in correctly predicting the known DR-altering and DR-neutral nsSNVs. They exhibited an overall accuracy of less than 50%, which was not better than random diagnosis. We found that the underlying problem is the markedly different evolutionary properties between positions harboring nsSNVs linked to drug responses and those observed for inherited diseases. To solve this problem, we developed a new diagnosis method, Drug-EvoD, which was trained on the evolutionary properties of nsSNVs associated with drug responses in a sparse learning framework. Drug-EvoD achieves a TPR of 84% and a TNR of 53%, with a balanced accuracy of 69%, which improves upon other methods significantly.

Conclusions

The new tool will enable researchers to computationally identify nsSNVs that may affect drug responses. However, much larger training and testing datasets are needed to develop more reliable and accurate tools.

Text mining for drug-drug interaction

In order to understand the mechanisms of drug-drug interaction (DDI), the study of pharmacokinetics (PK), pharmacodynamics (PD), and pharmacogenetics (PG) data are significant. In recent years, drug PK parameters, drug interaction parameters, and PG data have been unevenly collected in different databases and published extensively in literature. Also the lack of an appropriate PK ontology and a well-annotated PK corpus, which provide the background knowledge and the criteria of determining DDI, respectively, lead to the difficulty of developing DDI text mining tools for PK data collection from the literature and data integration from multiple databases.To conquer the issues, we constructed a comprehensive pharmacokinetics ontology. It includes all aspects of in vitro pharmacokinetics experiments, in vivo pharmacokinetics studies, as well as drug metabolism and transportation enzymes. Using our pharmacokinetics ontology, a PK corpus was constructed to present four classes of pharmacokinetics abstracts: in vivo pharmacokinetics studies, in vivo pharmacogenetic studies, in vivo drug interaction studies, and in vitro drug interaction studies. A novel hierarchical three-level annotation scheme was proposed and implemented to tag key terms, drug interaction sentences, and drug interaction pairs. The utility of the pharmacokinetics ontology was demonstrated by annotating three pharmacokinetics studies; and the utility of the PK corpus was demonstrated by a drug interaction extraction text mining analysis.The pharmacokinetics ontology annotates both in vitro pharmacokinetics experiments and in vivo pharmacokinetics studies. The PK corpus is a highly valuable resource for the text mining of pharmacokinetics parameters and drug interactions.

Importance of multi-p450 inhibition in drug-drug interactions: evaluation of incidence, inhibition magnitude, and prediction from in vitro data

Drugs that are mainly cleared by a single enzyme are considered more sensitive to drug-drug interactions (DDIs) than drugs cleared by multiple pathways. However, whether this is true when a drug cleared by multiple pathways is coadministered with an inhibitor of multiple P450 enzymes (multi-P450 inhibition) is not known. Mathematically, simultaneous equipotent inhibition of two elimination pathways that each contribute half of the drug clearance is equal to equipotent inhibition of a single pathway that clears the drug. However, simultaneous strong or moderate inhibition of two pathways by a single inhibitor is perceived as an unlikely scenario. The aim of this study was (i) to identify P450 inhibitors currently in clinical use that can inhibit more than one clearance pathway of an object drug in vivo and (ii) to evaluate the magnitude and predictability of DDIs caused by these multi-P450 inhibitors. Multi-P450 inhibitors were identified using the Metabolism and Transport Drug Interaction Database. A total of 38 multi-P450 inhibitors, defined as inhibitors that increased the AUC or decreased the clearance of probes of two or more P450s, were identified. Seventeen (45%) multi-P450 inhibitors were strong inhibitors of at least one P450, and an additional 12 (32%) were moderate inhibitors of one or more P450s. Only one inhibitor (fluvoxamine) was a strong inhibitor of more than one enzyme. Fifteen of the multi-P450 inhibitors also inhibit drug transporters in vivo, but such data are lacking on many of the inhibitors. Inhibition of multiple P450 enzymes by a single inhibitor resulted in significant (>2-fold) clinical DDIs with drugs that are cleared by multiple pathways such as imipramine and diazepam, while strong P450 inhibitors resulted in only weak DDIs with these object drugs. The magnitude of the DDIs between multi-P450 inhibitors and diazepam, imipramine, and omeprazole could be predicted using in vitro data with similar accuracy as probe substrate studies with the same inhibitors. The results of this study suggest that inhibition of multiple clearance pathways in vivo is clinically relevant, and the risk of DDIs with object drugs may be best evaluated in studies using multi-P450 inhibitors.