Richard A. Corley

Evaluation of Physiologically Based Models of Pregnancy and Lactation for Their Application in Children's Health Risk Assessments

In todays scientific and regulatory climates, an increased emphasis is placed on the potential health impacts for children exposed either in utero or by nursing to drugs of abuse, pharmaceuticals, and industrial or consumer chemicals. As a result, there is a renewed interest in the development and application of biologically based computational models that can be used to predict the dosimetry (or ultimately response) in a developing embryo, fetus, or newborn. However, fundamental differences between animal and human development can create many unique challenges. For example, unlike models designed for adults, biologically based models of pre- and postnatal development must deal with rapidly changing growth dynamics (maternal, embryonic, fetal, and neonatal), changes in the state of differentiation of developing tissues, uniquely expressed or uniquely functioning signal transduction or enzymatic pathways, and unusual routes of exposure (e.g., maternal-mediated placental transfer and lactation). In cases where these challenges are overcome or addressed, biological modeling will likely prove useful in assessments geared toward childrens health, given the contributions that this approach has already made in cancer and non-cancer human health risk assessments. Therefore, the purpose of this review is to critically evaluate the current state of the art in physiologically based pharmacokinetic (PBPK) and pharmacodynamic (PD) modeling of the developing embryo, fetus, or neonate and to recommend potential steps that could be taken to improve their use in childrens health risk assessments. The intent was not to recommend improvements to individual models per se, but to identify areas of research that could move the entire field forward. This analysis includes a brief summary of current risk assessment practices for developmental toxicity, with an overview of developmental biology as it relates to species-specific dosimetry. This summary should provide a general context for understanding the tension that exists in modeling between describing biological processes in exquisite detail vs. the simplifications that are necessary due to lack of data (or through a sensitivity analysis, determined to be of little impact) to develop individual PBPK or PD models. For each of the previously published models covered in this review, a description of the underlying assumptions and model structures as well as the data and methods used in model development and validation are highlighted. Although several of the models attempted to describe target tissues in the developing embryo, fetus, or neonate of laboratory animals, extrapolations to humans were largely limited to maternal blood or milk concentrations. Future areas of research therefore are recommended to extend the already significant progress that has been made in this field and perhaps address many of the technical, policy, and ethical issues surrounding various approaches for decreasing the uncertainty in extrapolating from animal models to human pregnancies or neonatal exposures.

Dosimetry considerations in the enhanced sensitivity of male Wistar rats to chronic ethylene glycol-induced nephrotoxicity.

Male Wistar rats have been shown to be the most sensitive sex, strain and species to ethylene glycol-induced nephrotoxicity in subchronic studies. A chronic toxicity and dosimetry study was therefore conducted in male Wistar rats administered ethylene glycol via the diet at 0, 50, 150, 300, or 400 mg/kg/day for up to twelve months. Subgroups of animals were included for metabolite analysis and renal clearance studies to provide a quantitative basis for extrapolating dose-response relationships from this sensitive animal model in human health risk assessments. Mortality occurred in 5 of 20 rats at 300 mg/kg/day (days 111-221) and 4 of 20 rats at 400 mg/kg/day (days 43-193), with remaining rats at this dose euthanized early (day 203) due to excessive weight loss. Increased water consumption and urine volume with decreased specific gravity occurred at 300 mg/kg/day presumably due to osmotic diuresis. Calculi (calcium oxalate crystals) occurred in the bladder or renal pelvis at > or =300 mg/kg/day. Rats dying early at > or =300 mg/kg/day had transitional cell hyperplasia with inflammation and hemorrhage of the bladder wall. Crystal nephropathy (basophilic foci, tubule or pelvic dilatation, birefringent crystals in the pelvic fornix, or transitional cell hyperplasia) affected most rats at 300 mg/kg/day, all at 400 mg/kg/day, but none at < or =150 mg/kg/day. No significant differences in kidney oxalate levels, the metabolite responsible for renal toxicity, were observed among control, 50 and 150 mg/kg/day groups. At 300 and 400 mg/kg/day, oxalate levels increased proportionally with the nephrotoxicity score supporting the oxalate crystal-induced nephrotoxicity mode of action. No treatment-related effects on the renal clearance of intravenously infused (3)H-inulin, a marker for glomerular filtration, and (14)C-oxalic acid were observed in rats surviving 12 months of exposure to ethylene glycol up to 300 mg/kg/day. In studies with naïve male Wistar and F344 rats (a less sensitive strain), a significant difference was observed in oxalate clearances between young rats (i.e. Wistar clearance < F344) but not in age-matched old rats. Regardless, the ratios of oxalate:inulin clearances in these two strains of rats, including those exposed to ethylene glycol, were all < 1, suggesting that a fraction of the filtered oxalate is reabsorbed. Other species, including humans, typically have clearance ratios >1 and are more effective at clearing oxalic acid by both glomerular filtration and active secretion. Thus, the lower renal clearance and kidney accumulation of oxalates in male Wistar rats enhances their sensitivity, which will be a factor in human risk assessments. The benchmark dose values (BMD05, BMDL05) were 170 mg/kg/day and 150 mg/kg/day for nephropathy, and 170 mg/kg/day and 160 mg/kg/day for birefringent crystals, using incidence times severity data in each case. The NOAEL of 150 mg/kg/day is the same as that reported after 16-week exposure and appears to be a threshold dose below which no renal toxicity occurs, regardless of exposure duration.

Magnetic resonance imaging and computational fluid dynamics (CFD) simulations of rabbit nasal airflows for the development of hybrid CFD/PBPK models

The percentages of total airflows over the nasal respiratory and olfactory epithelium of female rabbits were calculated from computational fluid dynamics (CFD) simulations of steady-state inhalation. These airflow calculations, along with nasal airway geometry determinations, are critical parameters for hybrid CFD/physiologically based pharmacokinetic models that describe the nasal dosimetry of water-soluble or reactive gases and vapors in rabbits. CFD simulations were based upon three-dimensional computational meshes derived from magnetic resonance images of three adult female New Zealand White (NZW) rabbits. In the anterior portion of the nose, the maxillary turbinates of rabbits are considerably more complex than comparable regions in rats, mice, monkeys, or humans. This leads to a greater surface area to volume ratio in this region and thus the potential for increased extraction of water soluble or reactive gases and vapors in the anterior portion of the nose compared to many other species. Although there was considerable interanimal variability in the fine structures of the nasal turbinates and airflows in the anterior portions of the nose, there was remarkable consistency between rabbits in the percentage of total inspired airflows that reached the ethmoid turbinate region (~50%) that is presumably lined with olfactory epithelium. These latter results (airflows reaching the ethmoid turbinate region) were higher than previous published estimates for the male F344 rat (19%) and human (7%). These differences in regional airflows can have significant implications in interspecies extrapolations of nasal dosimetry.

Extension of a PBPK model for ethylene glycol and glycolic acid to include the competitive formation and clearance of metabolites associated with kidney toxicity in rats and humans

A previously developed PBPK model for ethylene glycol and glycolic acid was extended to include glyoxylic acid, oxalic acid, and the precipitation of calcium oxalate that is associated with kidney toxicity in rats and humans. The development and evaluation of the PBPK model was based upon previously published pharmacokinetic studies coupled with measured blood and tissue partition coefficients and rates of in vitro metabolism of glyoxylic acid to oxalic acid, glycine and other metabolites using primary hepatocytes isolated from male Wistar rats and humans. Precipitation of oxalic acid with calcium in the kidneys was assumed to occur only at concentrations exceeding the thermodynamic solubility product for calcium oxalate. This solubility product can be affected by local concentrations of calcium and other ions that are expressed in the model using an ion activity product estimated from toxicity studies such that calcium oxalate precipitation would be minimal at dietary exposures below the NOAEL for kidney toxicity in the sensitive male Wistar rat. The resulting integrated PBPK predicts that bolus oral or dietary exposures to ethylene glycol would result in typically 1.4-1.6-fold higher peak oxalate levels and 1.6-2-fold higher AUCs for calcium oxalate in kidneys of humans as compared with comparably exposed male Wistar rats over a dose range of 1-1000 mg/kg. The converse (male Wistar rats predicted to have greater oxalate levels in the kidneys than humans) was found for inhalation exposures although no accumulation of calcium oxalate is predicted to occur until exposures are well in excess of the theoretical saturated vapor concentration of 200mg/m(3). While the current model is capable of such cross-species, dose, and route-of-exposure comparisons, it also highlights several areas of potential research that will improve confidence in such predictions, especially at low doses relevant for most human exposures.

A real-time methodology to evaluate the nasal absorption of volatile compounds in anesthetized animals

Nasal dosimetry models that combine computational fluid dynamics and physiologically based pharmacokinetic modeling incorporate information on species-specific anatomical differences, including nasal airflow, mucosal diffusion, clearance-extraction, and metabolism specific to different epithelial layers. As such, these hybrid models have the potential to improve interspecies dosimetric comparisons, and may ultimately reduce uncertainty associated with calculation of reference concentrations. Validation of these models, however, will require unique experimental data. To this end, a method for evaluating the uptake of a prototypical compound, methyl iodide (MeI), in the nasal cavity of the intact animal was developed. The procedure involved insertion of a small-diameter air-sampling probe in the depth of the nasal cavity to the nasopharynx region in anesthetized animals. The exterior portion of the probe was connected directly to a mass spectrometer to provide a continual real-time analysis of concentrations of MeI in the nasal cavity. A plethysmography system was used to monitor breathing parameters, including frequency and tidal volume for each animal. Animals were placed in a sealed glass chamber and exposed to MeI at initial chamber concentrations ranging from 1 to 50 ppm. Studies were conducted on nu2009=u20093 rabbits per exposure concentration for a total of nine animals and nu2009=u20096 rats at a single exposure concentration of 1 ppm. In the rabbit, the percent of MeI absorbed in the nasal cavity ranged from 57 to 92% (average 72u2009±u200911) regardless of exposure concentration. Similarly, the percent of MeI absorbed in the nasal cavity of the rat ranged from 51 to 71% (average 63u2009±u20098).

In vitro glutathione conjugation of methyl iodide in rat, rabbit, and human blood and tissues.

Methyl iodide (MeI) is an intermediate in the manufacture of some pesticides and pharmaceuticals, and is under review for US registration as a non-ozone depleting alternative for methyl bromide for pre-plant soil fumigation. MeI is primarily metabolized via conjugation with glutathione (GSH), with further metabolism to S-methyl cysteine and methanethiol. To facilitate extrapolations of animal pharmacokinetic data to humans, rate constants for the GSH metabolism of MeI were determined in cytosols prepared from the liver and kidneys of rats, human donors, female rabbits, and rabbit fetuses, from rabbit olfactory and respiratory epithelium, and from rabbit and rat blood using a headspace vial equilibration technique and two-compartment mathematical model. MeI was metabolized in liver and kidney from adults of all three species, but metabolism was not detectable in fetal rabbit kidney. Maximal metabolic rates (Vmax) were similar in liver from rat and human donors (~40 and 47u2009nmol/min/mg, respectively) whereas the Vmax rates in kidney cytosols varied approximately three-fold between the three species. No difference was observed in the loss of MeI from active and inactive whole blood from either rats or rabbits. The metabolism in olfactory and respiratory epithelial cytosol had Michaelis–Menten constant (Km) values that were several times higher than for any other tissue, suggesting essentially first-order metabolism in the nose. The metabolism of MeI in human liver cytosol prepared from five individual donors indicated two potential populations, one high affinity/low capacity and one with a lower affinity but higher capacity.

Studies supporting the development of a physiologically based pharmacokinetic (PBPK) model for methyl iodide: pharmacokinetics of sodium iodide (NaI) in pregnant rabbits

Methyl iodide (MeI) is a water soluble monohalomethane that is metabolized in vivo to release iodide (I−). A physiologically based pharmacokinetic (PBPK) model exists for iodide in adult rats, pregnant rats and fetuses, and lactating rats and neonates, but not for pregnant rabbits and fetuses, which have been used extensively for toxicity testing with MeI. Thus, this study was conducted to determine the blood and tissue distribution kinetics of radioiodide in pregnant rabbits and fetuses. Timed-pregnant New Zealand White rabbits received a single intravenous injection of the sodium salt of iodine-131 (Na131I) at either a high (10u2009mg/kg body weight) or low (0.75u2009mg/kg body weight) dose on gestation day 25. At various intervals ranging from 0.5 to 24u2009h post- injection, blood and tissues (thyroid, stomach contents, and skin) were collected from each doe, and blood, stomach contents, thyroid, trachea, and amniotic fluid were collected from a random sampling of three fetuses per doe per time point. Radioiodide accumulated as expected in the thyroid of maternal animals, where concentrations were the highest of any maternal tissues measured in both dose groups. Radioiodide also accumulated in fetal blood and tissues; levels were consistently higher than maternal levels and, unlike maternal tissues, showed no evidence of clearance over the 24-h sampling period. In contrast to observations in the maternal animals, fetal stomach contents showed the highest accumulation of radioiodide for both dose groups by 1–2u2009h after dosing, followed by the trachea and thyroid tissues, with the lowest concentrations of radioiodide in the amniotic fluid and blood. There was no evidence for preferential accumulation of radioiodide in fetal thyroid tissues.

Oral Reference Dose for ethylene glycol based on oxalate crystal-induced renal tubule degeneration as the critical effect

Several risk assessments have been conducted for ethylene glycol (EG). These assessments identified the kidney as the primary target organ for chronic effects. None of these assessments have incorporated the robust database of species-specific toxicokinetic and toxicodynamic studies with EG and its metabolites in defining uncertainty factors used in reference value derivation. Pertinent in vitro and in vivo studies related to one of these metabolites, calcium oxalate, and its role in crystal-induced nephropathy are summarized, and the weight of evidence to establish the mode of action for renal toxicity is reviewed. Previous risk assessments were based on chronic rat studies using a strain of rat that was later determined to be less sensitive to the toxic effects of EG. A recently published 12-month rat study using the more sensitive strain (Wistar) was selected to determine the point of departure for a new risk assessment. This approach incorporated toxicokinetic and toxicodynamic data and used Benchmark Dose methods to calculate a Human Equivalent Dose. Uncertainty factors were chosen, depending on the quality of the studies available, the extent of the database, and scientific judgment. The Reference Dose for long-term repeat oral exposure to EG was determined to be 15 mg/kg bw/d.

Development of a Zealand white rabbit deposition model to study inhalation anthrax.

Abstract Despite using rabbits in several inhalation exposure experiments to study diseases such as anthrax, there is a lack of understanding regarding deposition characteristics and fate of inhaled particles (bio-aerosols and viruses) in the respiratory tracts of rabbits. Such information allows dosimetric extrapolation to humans to inform human outcomes. The lung geometry of the New Zealand white rabbit (referred to simply as rabbits throughout the article) was constructed using recently acquired scanned images of the conducting airways of rabbits and available information on its acinar region. In addition, functional relationships were developed for the lung and breathing parameters of rabbits as a function of body weight. The lung geometry and breathing parameters were used to extend the existing deposition model for humans and several other species to rabbits. Evaluation of the deposition model for rabbits was made by comparing predictions with available measurements in the literature. Deposition predictions in the lungs of rabbits indicated smaller deposition fractions compared to those found in humans across various particle diameter ranges. The application of the deposition model for rabbits was demonstrated by extrapolating deposition predictions in rabbits to find equivalent human exposure concentrations assuming the same dose-response relationship between the two species. Human equivalent exposure concentration levels were found to be much smaller than those for rabbits.