Current Drug Metabolism - Volume 13, Issue 7, 2012

Human development of an individual from a fertilized ovum to maturity alters the body anatomy and physiology. Changes of size and function, from birth onwards, cause significant alterations in the pharmacokinetics (PK) of drugs and subsequently their response pharmacodynamics (PD) in infants and children from those in adults. During the last three decades, hundreds of mechanistic and clinical pharmacology studies have been conducted to investigate the age-mediated changes of absorption, distribution, metabolism and excretion processes of drugs, which subsequently affect the pharmacology response and the safety in pediatric patients compared to adults. The practice of determining pediatric dose based on simplistic scaling of an adult dose assuming linear relationship between postnatal age and body weight or surface area that may lead to under prediction of therapeutic dose or over prediction of the dose is now under scrutiny. By understanding the disposition mechanism of therapeutic agents thoroughly, their potential drug interactions and their PK/PD relationships can be better determined in pediatric populations. As such, dosing regimens can be estimated based on actual clearance and exposure and not just by simplistic scaling of an adult dose. Accurate prediction of clearance in pediatrics is so critical that extensive translational research is warranted to improve our ability to estimate safe and efficacious doses in different pediatric populations from retrospective clinical studies in adults. Biotherapeutics, proteins and peptides-based drugs, generally depend on absorption (A), distribution (D), metabolism (M), and excretion (E) in their disposition as small molecules, but the underlying mechanisms and potential drug interaction propensity can be very different. In this article, the factors that alter pediatric and adult PK parameters are compared, and pediatric and adult PK parameters and potential drug interactions for selected biotherapeutics are summarized. Moreover, challenges of studying therapeutic proteins and peptides in pediatrics are discussed.

Trastuzumab emtansine (T-DM1) is an antibody-drug conjugate in clinical development for the treatment of human epidermal growth factor receptor 2 (HER2)-positive cancers. Herein, we describe a series of studies to assess T-DM1 absorption, distribution, metabolism, and excretion (ADME) in rats as well as to assess human exposure to T-DM1 catabolites. Following administration of unlabeled and radiolabeled T-DM1 in female Sprague Dawley rats as a single dose, plasma, urine, bile and feces were assessed for mass balance, profiling and identification of catabolites. In rats, the major circulating species in plasma was T-DM1, while DM1 concentrations were low (1.08 to 15.6 ng/mL). The major catabolites found circulating in rat plasma were DM1, [N-maleimidomethyl] cyclohexane-1- carboxylate-DM1 (MCC-DM1), and Lysine-MCC-DM1. These catabolites identified in rats were also detected in plasma samples from patients with HER2-positive metastatic breast cancer who received single-agent T-DM1 (3.6 mg/kg every 3 weeks) in a phase 2 clinical study. There was no evidence of tissue accumulation in rats or catabolite accumulation in human plasma following multiple dosing. In rats, T-DM1 was distributed nonspecifically to the organs without accumulation. The major pathway of DM1-containing catabolite elimination in rats was the fecal/biliary route, with up to 80% of radioactivity recovered in the feces and 50% in the bile. The rat T-DM1 ADME profile is likely similar to the human profile, although there may be differences since trastuzumab does not bind the rat HER2- like receptor. Further research is necessary to more fully understand the T-DM1 ADME profile in humans.

Trastuzumab emtansine (T-DM1) is an antibody-drug conjugate comprised of trastuzumab and the cytotoxic agent DM1 (derivative of maytansine) linked by a stable linker N-succinimidyl 4-(N-maleimidomethyl) cyclohexane-1-carboxylate (SMCC). T-DM1 targets an epitope located at subdomain IV of human epidermal growth factor receptor 2 (HER2). Pertuzumab is a monoclonal antibody that targets an epitope located at subdomain II of HER2, distinct from the epitope recognized by T-DM1. The pharmacokinetics (PK), safety, and efficacy of T-DM1 combined with pertuzumab were studied in a phase 1b/2 trial in 67 patients with HER2-positive, locally advanced or metastatic breast cancer (MBC). The therapeutic protein-drug interaction (TP-DI) potential of T-DM1 plus pertuzumab was evaluated. The PK of T-DM1–related analytes and pertuzumab were compared with historical PK data. The results show that the exposure of T-DM1 and DM1, as estimated by noncompartmental analyses, was comparable with that reported by historical single-agent studies in patients with HER2-positive MBC. T-DM1 clearance and volume of distribution in the central compartment, as estimated by population PK analysis, were also comparable between this study and historical single-agent studies in patients with HER2-positive MBC. Summary statistics of pertuzumab trough and maximal exposure (concentrations at predose and 15–30 minutes after the end of infusion at cycle 1 and at steady state) were similar with those observed in a representative historical single-agent study with the same dosing regimen. The visual predictive check plot by population simulation further confirmed that T-DM1 did not alter pertuzumab PK. Based on these data and the PK and pharmacodynamic properties of T-DM1 and pertuzumab, the risk of TP-DI appears to be low when T-DM1 and pertuzumab are given together.

Inflammatory diseases such as rheumatoid arthritis and psoriasis are characterized by increases in circulating cytokines, which play an important role in modulation of the disease state. Several marketed bio-therapeutics target cytokines and act as effective treatment strategies. Previous in-vitro and in-vivo studies have suggested that cytokines may have both direct and indirect effects on drug metabolizing enzyme levels in the liver. Few studies have characterized models to evaluate the risk of potential drug interactions that might be mediated by changes in cytokine levels. In the present studies the potential of three cytokines (IL-2, IL-6 and TNF-α) to modulate gene expression and activity of the major human cytochrome P450 (CYP) enzymes (CYP1A2, 2B6, 2C9, 2C19, 2D6, and 3A4) in cryopreserved human hepatocytes (CHH) was investigated. Significant decreases in the activity of all 6 CYP isoforms occurred in hepatocytes incubated with TNF-α or IL-6 (17-85percnt; and 22-76% of untreated control values, respectively). TNF-α down-regulated the gene expression of CYP1A2, 2D6 and 3A4 only, whereas IL-6 down-regulated gene expression of all of the tested CYP isoforms except 2D6. IL-2 had only mild effects on CYP activity and mRNA levels of examined isoforms. In CHH exposed to TNF-α, changes in CYP activity were not always paralleled by gene expression alterations for three of the examined CYP isoforms. These studies highlight several potential pitfalls in using isolated human hepatocytes for determination of drug interactions by bio-therapeutics including lack of correlation of mRNA and activity measurements for some CYP isoforms when using single time point determinations, and appropriateness of the model for indirect acting cytokine and cytokine modulators.

Exposure to cytokines can down-regulate hepatic cytochrome P450 enzymes. Accordingly, relief of inflammation by cytokinetargeted drug therapy has the potential to up-regulate cytochrome P450s and thereby increase clearance of co-administered drugs. This study examined the effects of the inflammatory cytokine, interleukin 1β (IL-1β), and IL-1β/interleukin 6 (IL-6) combinations on drug metabolizing enzymes in human hepatocyte culture. Treatment of hepatocytes with IL-1β revealed suppression of mRNA expression of several clinically important cytochrome P450 isoenzymes, with EC50 values that differed by isoenzyme. Suppression of CYP1A2 activity by IL-1β could not be measured in 3 of 5 donors due to lack of response, and in the two remaining donors the average EC50 was 450 pg/mL. CYP3A activity had an EC50 of suppression of 416 ± 454 pg/mL. Measurable EC50s were obtained for all 5 donors for CYP2C8, 3A4, 3A5, 4A11 and IL-6R mRNA with fold differences which varied between 9.5-fold (CYP2C8) to 109-fold (CYP4A11). When hepatocytes were treated with IL-1β and IL-6 in combination at concentrations which ranged from 1-100 pg/mL, IL-6 was the main determinant of increases in acute phase response marker mRNA and of decreases in CYP3A4 mRNA. There was no synergy between IL-1β and IL-6 in the regulation of cytochrome P450 mRNA when dosed in combination, although the effects of the two cytokines in combination were additive in certain instances. These data indicate that IL-1β and IL-6 both suppress cytochrome P450 mRNA and enzyme levels in vitro and that, at similar physiologically-relevant concentrations in vitro, IL-6 is more potent than IL-1β.

We report here a comprehensive evaluation of the effects of culture duration on the gene expression of P450 isoforms, uptake transporters and efflux transporters in human hepatocyte cultured in the absence and presence of the prototypical proinflammatory cytokine, interleukin-6 (IL-6). Primary collagen-matrigel sandwich cultures of human hepatocytes were cultured in supplemented William's E medium containing 0, 0.1, 0.5 and 5 ng/mL of IL-6 for the time periods of 2, 6, 12, 24 and 48 hrs. Real-time PCR was performed to quantify gene expression of acute phase proteins (suppressor of cytokine signaling 3 (SOCS-3), c-reactive protein (CRP) and lipopolysaccharide (LPS)-binding proteins (LBP)); P450 isoforms (CYPs 1A2, 2B6, 2C8, 2C9, 2D6, 3A4, and 3A5), uptake transporters (SLC10A1, SLC22A1, SLC22A7, SLCO1B1, SLCO1B3, SLCO2B1) and efflux transporters (ABCB1, ABCB11, ABCC2, ABCC3, ABCC4, ABCG2). SOCS-3, CRP, and LBP were extensively induced by IL-6, with maximal induction observed at 2 (SOCS-3) and 12 hrs (CRP; LBP), demonstrating that the cultured human hepatocytes responded to IL-6 treatment. In the untreated group (control), gene expression of P450 isoforms and uptake transporters decreased while efflux transporters remained relatively stable or increased with cultured duration. IL-6 predominantly caused down regulations of the genes studied, with the most significant changes observed at different treatment durations, apparently related to the stability of the basal levels of gene expression. For instance, for genes with unstable expression, which would decrease rapidly in culture (e.g CYP3A4), the most definitive down regulatory effects were observed at a relatively early time point (e.g. 12 hrs). In contrast, a longer treatment duration (e.g. 48 hrs) was required for genes with relatively stable expression levels in culture (e.g. ABCB1). Based on our findings, evaluation of multiple treatment durations rather than single treatment duration is recommended for the evaluation of biotherapeutics in cultured human hepatocytes where down regulation is expected.

The increased development and availability of therapeutic proteins (TP) requires a greater understanding of drug-drug interactions (DDI) in a poly-pharmacy environment. Medications that are likely co-administered in various patient populations have to be considered in the context of direct pharmacokinetic properties of co-administered drugs as well as indirect effects due to disease modification. Anticipating and managing drug-drug interactions are an important part of the clinical development of both small and large molecules. Animal models may provide useful information in this regard with respect to human translatability. However, very limited work has been done to better understand the value of animal models for predicting the potential for TP DDI. Potential DDI mechanisms of TPs, with the exception of cytokine-mediated alteration of drug metabolizing enzymes and transporters, are less well known. This review discusses the potential value and specific challenges in the use of animal models for TP DDI evaluation.

Recent advances in genomic technologies have enabled the identification of thousands of genetic variations that are associated with hundreds of complex human diseases or traits in genome-wide association studies (GWAS). The large number of genetic loci uncovered for each disease or trait along with the difficulty in pinpointing the underlying genes and mechanisms further testify to the complexity of human pathophysiology. To alleviate the challenges of GWAS, systems biology approaches have been utilized to map the molecular mechanisms underlying complex human diseases/traits via the integration of genetic variation, functional genomics (such as genetics of gene expression), pathways, and molecular networks. Similar approaches have been applied to a spectrum of drug metabolizing enzymes to discover novel functional genetic variations that affect the expression or activities of these enzymes as well as to define the regulatory pathways/networks of genes involved in drug metabolism and toxicology in key human tissues. We envision that the increased coverage of functional genetic polymorphisms, the availability of drug metabolism-centered gene networks, and the maturing methodologies previously developed for understanding complex human diseases can be applied to pharmacogenomic and toxicogenomic studies to further our understanding of inter-individual variability in drug efficacy and toxicity and eventually help direct personalized medicine.

Genetic polymorphisms of drug transporters as well as drug metabolizing enzymes have been documented to play a significant role in patients' responses to medication. A key requirement for advancing personalized medicine is the ability to rapidly and conveniently test for patients' genetic polymorphisms. We have recently developed a rapid and cost-effective method for single nucleotide polymorphism (SNP) detection, named Smart Amplification Process (SmartAmp), which enables us to detect genetic polymorphisms or mutations in 30 to 45 min under isothermal conditions without the need for DNA isolation and PCR amplification. This article presents the SmartAmp-based detection of SNPs in the thiopurine S-methyltransferase gene as well as in the ATP-binding cassette (ABC) transporter ABCC4 and ABCG2 genes that are critically involved in drug-induced adverse reactions. The SmartAmp method is expected to provide a practical and cost-effective tool for pharmacogenomics-based personalized medicine.

Single nucleotide polymorphisms (SNPs) are the commonest genetic variant in the human genome and have been associated with inter-individual differences in drug response. Finding the causative SNPs underlying variations in drug response has been a cornerstone of personalized medicine. However, as there are over 19 million SNPs, the task of finding causative SNPs underlying differences in drug response using in vitro and in vivo methods can be intimidating. SNP related web resources can be invaluable in the search for SNPs relevant to drug response phenotypes as they represent relatively cheaper yet efficient ways of prioritizing relevant SNPs for further study. These resources serve as repositories of SNP information or contain in silico tools that can predict the functionality of a SNP. More sophisticated resources integrate the information repository function with the predictive function to create a one stop SNP resource for researchers. SNP related web resources can also aid researchers in planning and analyzing different types of genetic association studies by aiding in selecting SNPs for genotyping in these studies. The focus of this mini review is to outline the SNP related web resources that are available to researchers and how these resources may aid researchers studying SNP-drug response phenotype associations. Through efficient utilization of SNP related web resources, researchers will hopefully be able accelerate the pace of SNP related research in pharmacogenomics by identifying high risk SNP variants contributing to drug response as well as developing novel therapeutic targets based on understanding how SNPs alter drug response pathways.

Single nucleotide polymorphisms (SNPs) are the most frequent variants in many genes and are promising markers in relation to drug responses in pharmacogenomics studies. In this review, we emphasized the importance of the cheminformatic-related and SNPrelated resources and tools and how they can improve pharmacogenomics studies. Currently, many cheminformatic resources are well developed and provide much information on drug metabolism and targeting. In parallel, there are also many well established SNP-related resources that are able to provide the information related to SNP genotyping, tag SNPs and functional classification. However, cheminformatic and SNP resources have not, as yet, been well-integrated to provide a user-friendly platform for pharmacogenomics studies. This paper presents a brief overview of the many available public resources for cheminformatics (DrugBank, PharmGKB and other drugrelated databases) and SNPs (dbSNP, HapMap, SNP500Cancer, SNP-RFLPing 2 and other SNP tools) and points out the importance of integrating cheminformatic and SNP resources for the future of pharmacogenomics.

Personalized medicine has gained significant attention over the last decade as technologies for understanding biological differences between individuals have advanced dramatically. There are many potential benefits of personalized medicine including minimizing risk of drug toxicity, increasing benefit from drugs used, contributing to the sustainability of the healthcare system and facilitating drug discovery and development programs. Unfortunately there are also many barriers such as cost, complexity, high quality evidence requirements, and the need for further education that have limited the clinical translation of pharmacogenomic tests to date. Issues that need to be clarified are also considered, such as the regulatory evidence requirements for pharmacogenomic tests and the need for multiple pathways and for pharmacogenomic marker development. These issues surrounding personalized medicine are contextualized using three contemporary examples of pharmacogenetic tests involving drug metabolising enzymes: UDP glucuronosyltransferase 1A1 and irinotecan toxicity, cytochrome P450 2C19 and clopidogrel efficacy, and cytochrome P450 2C9 and warfarin dosing.

We have developed and standardized a novel technology, mutant-enriched liquidchip (MEL), for clinical detection of EGFR mutations. The MEL integrates a mutant-enriched PCR procedure with liquidchip technology for detections of EGFR exon 19 deletions and L858R mutation on both formalin-fixed and paraffin-embedded (FFPE) slides and plasma samples from patients with non-small cell lung cancer (NSCLC). The detection sensitivity was 0.1% of mutant DNA in the presence of its wild-type DNA. The cross-reaction rate was lower than 5%. To evaluate the MEL platform, the EGFR mutation status of 59 patients with advanced NSCLC treated with EGFRTKIs (Tyrosine Kinase Inhibitors) were tested on their FFPE samples. EGFR exon 19 deletions and L858R were detected in 21 patients (21/59) and 76.2% (16/21) of them had partial response to the EGFR-TKIs, while by sequencing method, only 4 (4/59) mutations were detected. Plasma samples from 627 patients with various stages of NSCLC were examined with the MEL and 22% of EGFR exon 19 deletions and L858R were detected. Furthermore, in patients with advanced disease there are more mutations detected in plasma samples than in patients with less advanced disease. In conclusion, the MEL is a sensitive, stable, and robust technology for detecting EGFR DNA mutations from both FFPE and plasma samples from patients with NSCLC and is now routinely used for clinical diagnosis.

With the high speed DNA sequencing of genome, databases of genome data continue to grow, and the understanding of genetic variation between individuals grows as well. Single nucleotide polymorphisms (SNPs), the most common type of genetic variation, are an increasingly important resource for understanding the structure and function of the human genome and become a valuable resource for investigating the genetic basis of disease. During the past years, in addition to experimental approaches to characterize specific variants, intense bioinformatics techniques were applied to understand the effects of these genetic changes. In the genetics studies, one intends to understand the molecular basis of disease, and computational methods are becoming increasingly important for SNPs selection, prediction and understanding the downstream effects of genetic variation. The review provides systematic information on the available resources and methods for SNPs discovery and analysis. We also report some new results on DNA sequence-based prediction of SNPs in human cytochrome P450, which serves as an example of computational methods to predict and discover SNPs. Additionally, annotation and prediction of functional SNPs, as well as a comprehensive list of existing tools and online recourses, are reviewed and described.

Human CYP2E1 accounts for almost 2% of total CYP enzymes in the liver cells, and plays a crucial role in the metabolism of small molecular weight compounds. This enzyme is associated with the nearly 6% metabolisms of the currently clinical drugs. However, it is found that CYP2E1 has a non-hyperbolic kinetic profile that can not be explained by the common Michaelis-Menten mechanism. Further studies show that the non-hyperbolic kinetic behaviors are associated with multiple substrate binding, which is also known as the cooperative binding properties. However, the detailed mechanism for the cooperative binding is not clear by now. In this paper, we summarized the experimental and theoretical studies on the cooperative binding mechanism. Based on the structural analysis, a second substrate binding site is confirmed in human CYP2E1, which is located neither in the region near Leu103, Leu210 and Phe478, nor far from the active site. Additionally, two important residues Thr303 and Phe478 are also identified to be the key factors in the cooperative binding on the short-range and long-range effects, respectively. The former plays a crucial role in the positioning of substrates and in proton delivery to the active site; the latter is located between the substrate access channel and the active site, and exhibits directly effects on substrate access or on substrate positioning in the active site. All these points can provide useful information for the cooperative binding in human CYP2E1, revealing the detailed mechanism for the non-hyperbolic kinetic behaviors.