Sami Haddad

Utility of physiologically based pharmacokinetic models to drug development and rational drug discovery candidate selection

The present paper proposes a modeling and simulation strategy for the prediction of pharmacokinetics (PK) of drug candidates by using currently available in silico and in vitro based prediction tools for absorption, distribution, metabolism and excretion (ADME). These methods can be used to estimate specific ADME parameters (such as rate and extent of absorption into portal vein, volume of distribution, metabolic clearance in the liver). They can also be part of a physiologically based pharmacokinetic (PBPK) model to simulate concentration-time profiles in tissues and plasma resulting from the overall PK after intravenous or oral administration. Since the ADME prediction tools are built only on commonly generated in silico and in vitro data, they can be applied already in early drug discovery, prior to any in vivo study. With the suggested methodology, the following advantages of the mechanistic PBPK modeling framework can now be utilized to explore potential clinical candidates already in drug discovery: (i) prediction of plasma (blood) and tissue PK of drug candidates prior to in vivo experiments, (ii) supporting a better mechanistic understanding of PK properties, as well as helping the development of more rationale PK-PD relationships from tissue kinetic data predicted, and hence facilitating a more rational decision during clinical candidate selection, and (iii) the extrapolation across species, routes of administration and dose levels.

Physiological Modeling of Age-Specific Changes in the Pharmacokinetics of Organic Chemicals in Children

Age-specific changes in the pharmacokinetics of chemicals are primarily due to differences in physiological and biochemical factors. For integrating the available information on the age-dependent changes in the physiological and biochemical factors, and for evaluating their combined influence on the pharmacokinetics of chemicals, physiologically based pharmacokinetic (PBPK) models are potentially useful. The objectives of this study were, therefore, (1) to assemble information on age-specific differences in physiological parameters such as alveolar ventilation rate, cardiac output, tissue volumes, tissue blood flow rates, and tissue composition for children of various age groups, and (2) to incorporate these data within a PBPK model for simulating the inhalation pharmacokinetics of a highly metabolized, volatile organic chemical (furan) in children of specific age groups (6, 10, and 14 yr old). The age-specific data on various physiological parameters were assembled following a review of the relevant literature and the hepatic metabolism rate of furan was set equal to the liver blood flow rate in adults and children. The blood:air and tissue:blood partition coefficients were calculated using molecular structure information along with the data on the blood and tissue composition (lipid and water contents) in children and adults. The PBPK model was used to simulate the pharmacokinetics of furan in adults and children (6, 10, and 14 yr old) exposed continuously for 30 h to 1 w g/L of this chemical in inhaled air. The model simulations suggest that, for the same exposure conditions, the blood concentration of furan is likely to be greater in children by a factor of 1.5 (at steady state) than in adults, and the maximal factor of adult-children differences in liver concentration of furan metabolite is about 1.25. The PBPK model framework developed in this study should be useful for predicting the adult-children differences in internal dose of chemicals for risk assessment applications.

Relative lipid content as the sole mechanistic determinant of the adipose tissue:blood partition coefficients of highly lipophilic organic chemicals

The adipose tissue:blood partition coefficient (PCat:b) refers to the ratio of chemical concentration or solubility in adipose tissue and blood. The solubility of a chemical in adipose tissue or whole blood is equal to the sum total of its solubility in lipid and water fractions of these matrices. For highly lipophilic organic chemicals (HLOCs, i.e., chemicals with log n-octanol:water partition coefficients (PCo:w) greater than four), their solubility in the water fractions of both tissue and blood is negligible, and therefore their solubility in lipid fractions of tissue and blood alone determines PCat:b. Since the numerical value representing chemical solubility in lipids is likely to be the same for both blood lipids and adipose tissue lipids, the PCat:b values should be hypothetically, equal to the ratio of lipid content of adipose tissue and blood. The objective of the present study was therefore to verify whether the PCat:bs of HLOCs (volatile organics, dioxins, PCBs, PBBs, DDT) are equal to the ratio of adipose tissue and blood lipid levels. The data on lipid content of rat and human blood and adipose tissues were obtained from the literature. The calculated tissue:blood lipid ratios were comparable to the human and rat PCat:b of volatile organic chemicals, dioxins, PCBs, PBBs and/or DDT obtained from the literature. These results then suggest that, regardless of the identity and PCo:w of HLOCs, their PCat:b is equal to the ratio of lipid in adipose tissues and blood.

CHARACTERIZATION OF AGE-RELATED CHANGES IN BODY WEIGHT AND ORGAN WEIGHTS FROM BIRTH TO ADOLESCENCE IN HUMANS

The pharmacokinetics and tissue dose of chemicals may differ among individuals of a population, particularly between adults and children. The adult-children differences in pharmacokinetics arise from age-related changes in the physiological, biochemical, and physicochemical determinants of uptake and disposition of chemicals. The objectives of this study were to review the published literature to assemble data on the human body weight and organ weights as a function of age (specifically between birth and 18 yr old) and to analyze these data, in order to develop regression equations for calculating body weight and organ weights of children using age as the dependent function. The specific organs/tissues for which the data on age-related weight were obtained and analyzed include blood, adipose tissues, liver, lungs, brains, heart, kidneys, spleen, the reproductive organs (male: prostate gland, seminal vesicle, testes, and epididymis; female: ovaries, uterus, and uterine tubes), glands (adrenal, pituitary, thymus, pancreas, and thyroid), bone marrow (total and red), intestinal tract, stomach, muscle, skin (epidermis and dermis), and skeleton. In both male and female children, the sum of these organs is systematically lower than the body weight, and this discrepancy may be resolved with the additional availability and consideration of data on hypodermis weight. The equations and data on body weight and organ weights presented in this article should be useful for constructing age-specific, physiologically based pharmacokinetic models for children.

Development of Physiologically Based Toxicokinetic Models for Improving the Human Indoor Exposure Assessment to Water Contaminants: Trichloroethylene and Trihalomethanes

Generally, ingestion is the only route of exposure that is considered in the risk assessment of drinking water contaminants. However, it is well known that a number of these contaminants are volatile and lipophilic and therefore highly susceptible to being absorbed through other routes, mainly inhalation and dermal. The objective of this study was to develop physiologically based human toxicokinetic (PBTK) models for trihalomethanes (THM) and trichloroethylene (TCE) that will facilitate (1) the estimation of internal exposure to these chemicals for various multimedia indoor exposure scenarios, and (2) consideration of the impact of biological variability in the estimation of internal doses. Five PBTK models describing absorption through ingestion, inhalation and skin were developed for these contaminants. Their concentrations in ambient air were estimated from their respective tap water concentrations and their physicochemical characteristics. Algebraic descriptions of the physiological parameters, varying as a function of age, gender and diverse anthropometric parameters, allow the prediction of the influence of interindividual variations on absorbed dose and internal dosimetry. Simulations for various scenarios were done for a typical human (i.e., 70 kg, 1.7 m) as well as for humans of both genders varying in age from 1 to 90 years. Simulations show that ingestion contributes to less than 50% of the total absorbed dose or metabolized dose for all chemicals. This contribution to internal dosimetry, such as maximal venous blood concentrations (Cmax) and the area under the venous blood concentration time curve (AUC), decreases markedly (e.g., as low as 0.9% of Cmax for bromodichloromethane). The importance of this contribution varies mainly as a function of shower duration. Moreover, model simulations indicate that multimedia exposure is more elevated in children than adults (i.e., up to 200% of the adult internal dose). The models developed in this study allow characterization of the influence of the different routes of exposure and an improved estimation of the realistic multimedia exposure to volatile organic chemicals present in drinking water. Hence, such models will greatly improve health risk assessment for these chemicals.

In Vitro–In Vivo Extrapolation of Clearance: Modeling Hepatic Metabolic Clearance of Highly Bound Drugs and Comparative Assessment with Existing Calculation Methods

In vitro-in vivo extrapolation (IVIVE) is an important method for estimating the hepatic metabolic clearance (CL) of drugs. This study highlights a problematic area observed when using microsomal data to predict in vivo CL of drugs that are highly bound to plasma proteins, and further explores mechanisms for human CL predictions by associating additional processes to IVIVE disconnect. Therefore, this study attempts to develop a novel IVIVE calculation method, which consists of adjusting the binding terms in a well-stirred liver model. A comparative assessment between the IVIVE method proposed here and previously published methods of Obach (1999. Drug Metab Dispos 27:1350-1359) and Berezhkovskiy (2010. J Pharm Sci 100:1167-1783) was also performed. The assessment was confined by the availability of measured in vitro and in vivo data in humans for 25 drugs highly bound to plasma proteins, for which it can be assumed that metabolism is the major route of elimination. Here, we argue that a difference in drug ionization and binding proteins such as albumin (AL) and alpha-1-acid glycoprotein (AAG) in plasma and liver also needs to be considered in IVIVE based on mechanistic studies. Therefore, converting unbound fraction in plasma to liver essentially increased the predicted CL values, which resulted in much more accurate estimates of in vivo CL as compared with the other IVIVE methods tested. The impact on CL estimate was more apparent for drugs binding to AL than to AAG. This is a mechanistic rational for explaining a considerable proportion of the divergence between previously estimated and observed CL values. Human CL was predicted within 1.5-fold, twofold, and threefold of the observed CL for 84%, 96%, and 100% of the compounds, respectively. Overall, this study demonstrates a significant improvement in the mechanism-based prediction of metabolic CL for these 25 highly bound drugs from in vitro data determined with microsomes, which should facilitate the application of physiologically based pharmacokinetic (PBPK) models in drug discovery and development.

A physiologically based pharmacokinetic model for the assessment of infant exposure to persistent organic pollutants in epidemiologic studies.

Background It has been suggested that pre- and postnatal exposure to persistent organic pollutants (POPs) can promote several adverse effects in children, such as altered neurodevelopment. Epidemiologic studies to date have relied on the analysis of biological samples drawn pre- or post-natally for exposure assessment, an approach that might not capture some key events in the toxicokinetics of POPs. Objectives We aimed to build a generic physiologically based pharmacokinetic (PBPK) modeling framework for neutral POPs to assess infant toxicokinetic profiles and to validate the model using data on POP levels measured in mothers and infants from a Northern Québec Inuit population. Methods The PBPK model developed herein was based upon a previously published model to which an infant submodel was added. Using the model and maternal blood levels at the time of delivery, exposure to 1,1-dichloro-2,2-bis(p-chlorophenyl)ethylene (p,p′-DDE), 1,1,1-trichloro-2,2-bis(p-chlorophenyl)ethane (p,p′-DDT), hexachlorobenzene (HCB), β-hexachlorocyclohexane (β-HCH), 2,2′,3,4,4′,5′-hexachlorobiphenyl (PCB-138), 2,2′,4,4′,5,5′-hexachlorobiphenyl (PCB-153), and 2,2′,3,4,4′,5,5′-heptachlorobiphenyl (PCB-180) in mothers was estimated to subsequently simulate infant blood, breast milk, and cord blood POP concentration. Simulations were then compared with corresponding measured levels through Spearman correlation analyses. Results Predictions were highly correlated with measured concentrations for PCB-153, PCB-180, PCB-138, HCB, and p,p′-DDE (r = 0.83–0.96). Weaker correlations were observed for p,p′-DDT and β-HCH for which levels were near the limits of detection. Conclusion This is the first study to validate a PBPK model of POPs in infants on an individual basis. This approach will reduce sampling efforts and enable the use of individualized POP toxicokinetic profiles in the epidemiologic studies of POP adverse effects on child development.

Physiologically based pharmacokinetic modeling of persistent organic pollutants for lifetime exposure assessment: a new tool in breast cancer epidemiologic studies.

Background Despite experimental evidence, most epidemiologic studies to date have not supported an association between exposure to persistent organic pollutants (POP) and breast cancer incidence in humans. This may be attributable to difficulties in estimating blood/tissue POP concentration at critical time periods of carcinogenesis. Objectives In this work we aimed to develop a tool to estimate lifetime POP blood/tissue exposure and levels during any hypothesized time window of susceptibility in breast cancer development. Methods We developed a physiologically based pharmacokinetic (PBPK) model that can account for any given physiologic lifetime history. Using data on pregnancies, height, weight, and age, the model estimates the values of physiologic parameters (e.g., organ volume, composition, and blood flow) throughout a woman’s entire life. We assessed the lifetime toxicokinetic profile (LTP) for various exposure scenarios and physiologic factors (i.e., breast-feeding, growth, pregnancy, lactation, and weight changes). Results Simulations for three POPs [hexachlorobenzene, polychlorinated biphenyl (PCB)-153, PCB-180] using different lifetime physiologic profiles showed that the same blood concentration at 55 years of age can be reached despite totally different LTP. Aside from exposure levels, lactation periods and weight profile history were shown to be the factors that had the greatest impact on the LTP. Conclusions This new lifetime PBPK model, which showed the limitations of using a single sample value obtained around the time of diagnosis for lifetime exposure assessment, will enable researchers conducting environmental epidemiology studies to reduce uncertainty linked to past POP exposure estimation and to consider exposure during time windows that are hypothesized to be mechanistically critical in carcinogenesis.

Physiologically-Based Pharmacokinetic (PBPK) Models in Toxicity Testing and Risk Assessment

Physiologically-based pharmacokinetic (PBPK) modeling offers a scientifically-sound framework for integrating mechanistic data on absorption, distribution, metabolism and elimination to predict the time-course of parent chemical, metabolite(s) or biomarkers in the exposed organism. A major advantage of PBPK models is their ability to forecast the impact of specific mechanistic processes and determinants on the tissue dose. In this regard, they facilitate integration of data obtained with in vitro and in silico methods, for making predictions of the tissue dosimetry in the whole animal, thus reducing and/or refining the use of animals in pharmacokinetic and toxicity studies. This chapter presents the principles and practice of PBPK modeling, as well as the application of these models in toxicity testing and health risk assessments.

Toxicokinetic Modeling of Persistent Organic Pollutant Levels in Blood from Birth to 45 Months of Age in Longitudinal Birth Cohort Studies

Background: Despite experimental evidence that lactational exposure to persistent organic pollutants (POPs) can impact health, results from epidemiologic studies are inconclusive. Inconsistency across studies may reflect the inability of current methods to estimate children’s blood levels during specific periods of susceptibility. Objectives: We developed a toxicokinetic model to simulate blood POP levels in children from two longitudinal birth cohorts and aimed to validate it against blood levels measured at 6, 16, and 45 months of age. Methods: The model consisted of a maternal and a child lipid compartment connected through placental diffusion and breastfeeding. Simulations were carried out based on individual physiologic parameters; duration of breastfeeding; and levels of POPs measured in maternal blood at delivery, cord blood, or breast milk. Model validity was assessed through regression analyses of simulated against measured blood levels. Results: Simulated levels explained between 10% and 83% of measured blood levels depending on the cohort, the compound, the sample used to simulate children’s blood levels, and child’s age when blood levels were measured. Model accuracy was highest for estimated blood POP levels at 6 months based on maternal or cord blood levels. However, loss in model precision between the 6th and the 45th month was small for most compounds. Conclusions: Our validated toxicokinetic model can be used to estimate children’s blood POP levels in early to mid-childhood. Estimates can be used in epidemiologic studies to evaluate the impact of exposure during hypothesized postnatal periods of susceptibility on health.