Evaluation of the Potential Impact of Age- and Gender-Specific Pharmacokinetic Differences on Tissue Dosimetry

Harvey J. Clewell*,1, P. Robinan Gentry*, Tammie R. Covington*, Ramesh Sarangapani{dagger},2 and Justin G. Teeguarden*

* ENVIRON Health Sciences Institute, Ruston, Louisiana 71270; and {dagger} K. S. Crump Group, Inc., ICF Consulting, Research Triangle Park,North Carolina 27709

Received June 23, 2003; accepted February 12, 2004


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUPPLEMENTARY DATA
 REFERENCES
 
The physiological and biochemical processes that determine the tissue concentration time courses (pharmacokinetics) of xenobiotics vary, in some cases significantly, with age and gender. While it is known that age- and gender-specific differences have the potential to affect tissue concentrations and, hence, individual risk, the relative importance of the contributing processes and the quantitative impact of these differences for various life stages are not well characterized. The objective of this study was to identify age- and gender-specific differences in physiological and biochemical processes that affect tissue dosimetry and integrate them into a predictive physiologically based pharmacokinetic (PBPK) life-stage model. The life-stage model was exercised for several environmental chemicals with a variety of physicochemical, biochemical, and mode-of-action properties. In general, predictions of average pharmacokinetic dose metrics for a chemical across life stages were within a factor of two, although larger transient variations were predicted, particularly during the neonatal period. The most important age-dependent pharmacokinetic factor appears to be the potential for decreased clearance of a toxic chemical in the perinatal period due to the immaturity of many metabolic enzyme systems, although this same factor may also reduce the production of a reactive metabolite. Given the potential for age-dependent pharmacodynamic factors during early life, there may be chemicals and health outcomes for which decreased clearance over a relatively brief period could have a substantial impact on risk.

Key Words: pharmacokinetics; PBPK; children; age; gender; dosimetry.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUPPLEMENTARY DATA
 REFERENCES
 
Several recent initiatives by the U. S. Environmental Protection Agency (USEPA) have increased the emphasis on research to investigate age-specific risk issues. One such initiative, the Voluntary Children's Chemical Evaluation Program (VCCEP), is designed to provide data to improve understanding of potential health risks to children associated with chemical exposures. This program also provides for the development of methods specific to exposure assessment in children. In 2002, USEPA also announced the first coordinated effort by the agency to examine and prioritize environmental health threats to the elderly. In order to adequately address the concerns raised in these initiatives, it will be necessary to make optimal use of the resources available to obtain toxicological data on the early life period. Prioritization of toxicity study requirements can be aided by an understanding of age-related differences in pharmacokinetics and pharmacodynamics and their potential impact on the internal exposure to environmental chemicals. Methods for incorporating this pharmacokinetic information into the risk assessment process are also needed in order to assess whether exposure to a given chemical may be of greater concern during a particular life stage.

In response to this heightened interest in estimating age-related risks, a comprehensive review was conducted to identify the available quantitative information in humans related to age- and gender-dependent differences in physiological, biochemical, and pharmacokinetic parameters that may impact risk from chemical exposure (Clewell et al., 2002Go). This information was reviewed from a risk assessment perspective, and the key factors that are likely to have a significant impact on susceptibility, as it relates to estimates of target tissue exposure, were identified. Much of the available data were obtained from studies with pharmaceutical agents. The majority of the differences in pharmacokinetics identified were between neonates/children and adults, with fewer differences identified between young adults and the elderly. The results of this study are consistent with the results of studies that have been conducted to evaluate the pharmacokinetic differences between children and adults (Ginsberg et al., 2002Go; Hattis et al., 2003Go; Renwick et al., 2000Go).

However, in all of these studies it has been necessary to rely upon pharmacokinetic data for pharmaceuticals, which are more readily available than similar data for environmentally relevant chemicals. Unfortunately, the application of the results for pharmaceuticals to environmentally relevant compounds is problematic, due to the significant differences in the typical physicochemical and biochemical properties of pharmaceutical and environmental compounds. Pharmaceuticals tend to be water soluble, while environmental chemicals of concern are frequently lipophilic. The two classes of chemicals also tend to be substrates for different, although overlapping, subsets of the metabolic enzyme systems. For example, much of the information on age-dependent metabolic clearance of pharmaceuticals is for CYP3A4 substrates (Ginsberg et al., 2002Go), but this isozyme has not frequently been associated with the metabolism of environmental contaminants. On the other hand, CYP2E1 is associated with the toxicity of many environmental contaminants (Guengerich et al., 1991Go), but was not identified as a common drug metabolizing isozyme.

The primary objective of this study was to identify age- and gender-specific differences in physiological and biochemical processes that affect tissue dosimetry, and integrate them into a predictive physiologically based pharmacokinetic (PBPK) life-stage model that could be exercised for environmental chemicals with a variety of physicochemical, biochemical and mode-of-action properties in order to determine the interaction between the pharmacokinetic processes and the properties of the chemical.

PBPK modeling has been routinely used in risk assessment when extrapolating across route and species. The same qualities that make PBPK modeling attractive for these extrapolations also make it a useful platform for predicting age-dependent pharmacokinetics. Specifically, a PBPK model can provide a quantitative structure for incorporating into the risk assessment process information on the various age- and gender-specific pharmacokinetic factors that can impact the relationship between the external (environmental) exposure and the internal (biologically effective) target tissue exposure. Recent guidance from the International Programme on Chemical Safety (IPCS, 2001Go) provides an approach for replacing default uncertainty factors with chemical-specific adjustment factors (CSAFs). This approach divides the animal-to-human interspecies and human intraspecies uncertainty factors into toxicokinetic and toxicodynamic components, each of which can be replaced by a CSAF if data are available. For example, the magnitude of the factor for human variability in toxicokinetics (HKAF) may be calculated based on an evaluation of human variability in the area under the tissue concentration-time curve (AUC) or clearance. A PBPK model provides an excellent basis for performing such an evaluation (Lipscomb et al., 2004Go; Pelekis et al., 2001Go). This paper describes the initial development of a generic PBPK model that can be used to estimate age-specific CSAFs for any chemical, exposure route, and life stage, dependent only on the availability of adequate chemical-specific partitioning and metabolism information. Child–adult CSAFs predicted with such model could in principle be used in the context of the VCCEP initiative to replace or inform a child-specific uncertainty factor, and the same approach could also be applied to obtain CSAFs for other life stages, such as the elderly.

The purpose of the study presented here is to provide a proof-of-principle demonstration of the potential of PBPK modeling to address age-specific dosimetry issues quantitatively. It is not the intent of the authors to suggest that the predictions of this initial model for specific chemicals should be used quantitatively in support of risk assessments. The chemical-specific parameterization of this initial model is based on previously published PBPK descriptions for young adults. Predictions of the model for other life stages, while based on reasonable physiological and biochemical principles, should be considered to represent no more than exploratory dosimetry estimates. Further age-specific parameter refinement and chemical-specific model validation will be required before PBPK-based age extrapolation can be performed with confidence.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUPPLEMENTARY DATA
 REFERENCES
 
Selection of chemical class surrogates.
Six chemical classes with different physicochemical properties were targeted, and surrogate chemicals were selected to represent each class. The surrogates selected were based on the availability of chemical-specific information critical for PBPK modeling: metabolic parameters and partition coefficients. Examples were conducted for two nonvolatile or semivolatile classes and four volatile classes (Table 1). For the nonvolatile classes, one example was selected to represent highly lipophilic, nonvolatile compounds, for which 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) served as the surrogate. Nicotine was selected as a representative for water-soluble, nonvolatile or semivolatile chemicals.


View this table:
[in this window]
[in a new window]
 
TABLE 1 List of Surrogate Chemicals and Important Characteristics

 
The chemicals selected to serve as surrogates for the volatile classes represented not only differences in physicochemical properties, but also differences in metabolic characteristics, including metabolic production of stable metabolites or reactive intermediates, as well as single or competing metabolic pathways. Four volatile chemicals were selected with decreasing water solubility and increasing lipophilicity: isopropanol, vinyl chloride, methylene chloride, and perchloroethylene (PERC). Of these volatiles, vinyl chloride produces reactive intermediates via a single oxidative metabolic pathway, while methylene chloride has two competing metabolic pathways, oxidation and glutathione (GSH) conjugation, that both produce reactive intermediates. Isopropanol and perchloroethylene are both metabolized to stable compounds: acetone and trichloroacetic acid (TCA), respectively.

Model structure and parameters.
The life-stage PBPK model used in this study is an elaboration of a previously published adult model for isopropanol (Clewell et al., 2001bGo) (Fig. 1). The primary feature of the life-stage model is that it allows for the simulation of the time-dependence of physiological and biochemical parameters due to growth and aging. All physiological and biochemical parameters in the model were allowed to change with time, using either equations or interpolation between discrete data points, based on available age-related quantitative information, with only the chemical-specific parameters remaining constant. Since the adult model has been documented in previous publications (Clewell et al., 2001bGo; Gentry et al., 2002Go), only the modifications to the model necessary to support age-dependent simulations will be discussed. A description of the age- and gender-related information used to describe the physiological and biochemical parameters in the life-stage model is provided in the following sections, and the model equations are provided in the supplementary data, Appendix A. The model was coded using Advanced Continuous Simulation Language (ACSL, Aegis Technologies Group, Inc., Huntsville, AL).



View larger version (26K):
[in this window]
[in a new window]
 
FIG. 1. Schematic of the life-stage PBPK model.

 
If there were sufficient data, an equation was used in the model to describe the age- or gender-specific changes in a parameter; otherwise, linear interpolation between available data points was used to define the time evolution of the parameter. Where data were available for males and females, separate equations or data points were used to describe each gender. Equations were used to describe age-dependent changes in body weight, skin surface area, cardiac output, rapidly perfused tissue volume, and glomerular filtration rate (Supplementary Data, Appendix A). For some of these parameters, multiple equations were necessary to adequately describe the changes observed in the parameter throughout the entire period of life. Data on alveolar ventilation, metabolic capacity, and brain, liver, and fat weight were used for interpolation (Supplementary Data, Tables A, B, C, and D). As a check of the resulting model's structural framework, the predictions of the age-dependent model for isopropanol were compared with the predictions of the validated adult human PBPK model on which it was based (Clewell et al., 2001bGo). This comparison was performed at several ages, using the appropriate age-specific parameters in each model, to assure that the age-dependent model could accurately reproduce the output of the original model for the same life stage. However, these comparisons do not serve as validation of the predictions of the model for early life stages; the predictions of the published isopropanol model have only been validated against human kinetic data for the adult. The results of this comparison are shown in Figures 2A and 5A.



View larger version (30K):
[in this window]
[in a new window]
 
FIG. 2. Blood concentrations as a function of age for continuous lifetime oral exposure at a constant daily intake of 1 µg/kg/d for (A) isopropanol (IPA) and its metabolite acetone, and (D) perchloroethylene (PERC) and its primary metabolite TCA. Blood concentration and the rate of metabolite production per kg of liver as a function of age for continuous oral exposure at a constant daily intake of 1 µg/kg/d for (B) vinyl chloride (VC) and (C) methylene chloride (MC). Circles and squares in panel A indicate the predictions of the published IPA model (Clewell et al., 2001aGo) using the corresponding age-specific parameters.

 


View larger version (19K):
[in this window]
[in a new window]
 
FIG. 5. Blood concentrations of isopropanol (IPA) and its metabolite acetone as a function of age for (A) continuous inhalation exposure at a concentration of 1 ppb and (B) continuous dermal exposure of 18.5% of the total skin surface area to isopropanol in water at a concentration of 70 µg/l. Circles and squares in panel A indicate the predictions of the published IPA model (Clewell et al., 2001aGo) using the corresponding age-specific parameters.

 
Physiological parameters.
Information on monthly body weights for birth and age 1 month were obtained from USEPA (1997)Go, and weights for age 2 months to 90 years were obtained from NHANES (1995)Go. A separate equation was needed for each of four different time periods (three Gompertz equations and one cubic equation), and the equations were fit to the data such that the curve was continuous. Separate sets of equations were developed to fit the data for males and females (Supplementary Data, Figure A).

Total surface areas (cm2) corresponding to age-specific body weights were obtained from nomograms provided in ICRP (1975)Go. Two separate cubic equations were necessary to adequately describe the data (Supplementary Data, Figure B). Because the equation is a function of body weight, the same equation was used to describe age-related changes in surface area for both males and females.

Data describing pulmonary ventilation (m3/day) for various ages are provided in USEPA (1997)Go. These values were converted to alveolar ventilation rates for use in the model, based on the assumption that alveolar ventilation is approximately two-thirds of pulmonary ventilation. The values reported in USEPA (1997)Go, which consider typical activity levels at different ages, were matched to age-specific body weights calculated as above. Because the resulting ventilation rates did not follow a smooth curve, alveolar ventilation is described as a function of time, in months, using linear interpolation between discrete data points (Supplementary Data, Table A).

Data on the relationship of cardiac output and pulmonary ventilation to level of activity (oxygen consumption) were taken from Åstrand (1983)Go (Supplementary Data, Figure C). For various values of oxygen uptake, corresponding values for cardiac output and pulmonary ventilation rate were extracted from these figures. As with the USEPA (1997)Go ventilation rates, these rates were adjusted by a ratio of 2/3 to obtain alveolar ventilation rates for use in the model. Based on the resulting relationship, cardiac output was described as a function of alveolar ventilation using a Gompertz curve (Supplementary Data, Figure D).

Ogiu et al. (1997)Go provided age-related data on brain and liver weights as fractions of body weight for a Japanese population. Data on fat content as a fraction of body weight through age 20 was obtained from Hattis (2003)Go, and was combined with data from ICRP (1975)Go. These data were used in the model to calculate fractional tissue volumes as a function of time in months (Supplementary Data, Table B). These calculated fractional values were then multiplied by body weight to get the necessary tissue volumes.

Data on intestinal weight by age was used to estimate relative age-dependence of the volume of GI tract tissue (ICRP, 1975Go). Three separate Gompertz equations were used to describe the available data (Supplementary Data, Figure E). Due to a lack of data on adults, it was assumed that the fractional volume of the GI tract remained constant after age 22.

As in most simplified PBPK descriptions, the rapidly perfused tissue compartment in the life-stage model is a loosely defined composite of a large number of tissues, most of which represent a very small fraction of the body weight. Age-dependent data were not available for many of these tissues. Since the volume of the GI tract is by far the largest component of the rapidly perfused tissue volume, data on the weight of this tissue (ICRP, 1975Go) were used as a surrogate for the age-dependent changes in this lumped compartment. The age-dependent equation describing fractional GI tract volume was multiplied by a constant to produce a fractional volume of rapidly perfused tissues at age 25 that was consistent with the previously published adult model (Clewell et al., 2001bGo). The skin volume was modeled in the same manner as in the published adult model (Clewell et al., 2001bGo).

The volume of slowly perfused tissues was modeled as 84% of the total body weight minus the volume of the other tissues. This assures that the total weight of perfused tissues for which the model accounts is 84% of the body weight (Clewell et al., 2001bGo). The rest of the body (16%) is assumed to be nonperfused tissue (cortical bone, hair, gut contents, etc.).

Due to the lack of data on tissue blood flow changes with age, the blood flows were assumed to change proportionally with the tissue volumes. The adult fractional tissue blood flows from the published adult model (Clewell et al., 2001bGo) were used along with the age-specific tissue volumes and the tissue volumes at age 25 years. To maintain mass balance for the blood flows, the age-specific fractional blood flows were normalized to always sum to unity.

Metabolic parameters.
Quantitative information on the age-related changes in the relevant metabolic pathways was used in modeling each of the surrogate chemicals (Supplementary Data, Table C). For the oxidative metabolism of vinyl chloride, methylene chloride, and perchloroethylene, data on the development of CYP2E1 were used (Vieira et al., 1996Go). For the metabolism of TCDD, data for CYP1A2 (Sonnier and Cresteil, 1998Go) were used, while for isopropanol, both alcohol dehydrogenase (ADH) data (Pikkarainen and Räihä, 1967Go) and CYP2E1 data (for acetone metabolism) were used. No information on age-related differences was available for the CYP2A6 pathway, which is the primary metabolic pathway for nicotine; therefore, age-related information available on the development of CYP2C (Treluyer et al., 1997Go) was used as a surrogate for the purposes of this case study. Only limited quantitative information is available on the development of the glutathione-S-transferase (GST) pathway. Based on these limited data (Mendrala et al., 1993Go; Pacifici et al., 1981Go), the trend associated with the development of this pathway is similar to that observed for ADH. Therefore, age-related information on the metabolic capacity for ADH was used as a surrogate to predict age-related changes in GST for methylene chloride.

Due to the lack of gender-specific metabolism data, the same data were used for both males and females. Metabolism by all enzyme systems was initiated at zero at birth, with the exception of ADH, for which data were available that demonstrated prenatal activity; available data for the other metabolic pathways indicate no fetal metabolic activity (Pikkarainen and Räihä, 1967Go; Sonnier and Cresteil, 1998Go; Vieira et al., 1996Go). For ADH, linear interpolation between the last prenatal value and the first postnatal value was used to estimate a value for ADH at birth.

Age-specific metabolism rates were calculated using the adult metabolism rate (mg/hr), the adult liver volume (kg), the age-specific liver volume (kg), and the appropriate linearly interpolated fractional activity. Scenarios were also evaluated in which metabolism was artificially delayed for the first 6 months of life in order to evaluate the potential impact of late development of an enzyme system. In these cases, enzyme activity was assumed to be zero until 6 months of age, after which it was assumed to increase linearly from zero to the first fractional value available after 6 months.

Urinary clearance parameters.
Data on glomerular filtration rate (GFR) at different ages were obtained from several sources (Braunlich, 1977Go; Milsap and Jusko, 1994Go; Plunkett et al., 1992Go). It was determined that, as a function of age, GFR is best described in several distinct phases. GFR ranges from 0.12 to 0.24 l/hr at birth, and then increases by a factor of about 4 within 72 hours (Milsap and Jusko, 1994Go). Between days 3 and 10 of life, GFR continues to increase until the normalized value is approximately that of an adult (Braunlich, 1977Go; Milsap and Jusko, 1994Go; Plunkett et al., 1992Go). After reaching normalized adult values, GFR scales as a constant function (per 1.73 m2) of total body surface area until approximately 30 years of age, after which GFR declines from the adult value at 30 years of age by 0.66% per year. These phases were modeled using five separate equations (Supplementary Data, Appendix A). The first equation sets GFR to be 0.12 l/hr until 1 day of age. The second equation increases GFR linearly by a factor of four over the next 3 days, and the third equation linearly increases GFR to normalized adult levels by the end of 10 days of age. The fourth equation was developed using data on body weight and total body surface area and calculates GFR as a function (per 1.73 m2) of total body surface area until 30 years of age. The last equation decreases GFR from the adult value at 30 years of age by 0.66% each year.

Renal blood flow data were available for ages up to 50 days (Braunlich, 1977Go), (Supplementary Data, Figure F) and for ages 25 to 85 years (Mayersohn, 1994Go) (Supplementary Data, Table D). The data from Braunlich (1977)Go indicated that, by 50 days of age, fractional renal blood flow has reached adult levels. Age-specific urinary clearance rates were calculated using the adult urinary clearance rate (l/hr) and the ratio of the linearly interpolated age-specific GFR or renal blood flow to the corresponding adult GFR or renal blood flow.

Model simulations for surrogate chemicals.
The conditions of the model simulations are summarized in Table 2. For each surrogate chemical, continuous lifetime oral exposure was simulated (birth to 75 years) for both males and females at a daily dose rate of 1 µg/kg/d (with the exception of TCDD, for which a daily dose of 1 ng/kg/d was used). For TCDD, the gestational period was also simulated to account for bioaccumulation prior to birth; the rapid clearance of the other surrogate chemicals makes this additional step unnecessary. A comparison of in utero and neonatal exposure with these same chemicals was performed in a separate study (Gentry et al., 2003Go). Oral exposure was modeled using a constant intake rate, rather than attempting to simulate a time-varying diurnal ingestion pattern, since the impact of widely varying ingestion patterns on average daily internal dose metrics, such as those used in this analysis, has been shown to be relatively small (NRC, 1986Go).


View this table:
[in this window]
[in a new window]
 
TABLE 2 Age- and Chemical-Specific Information for Model Simulations

 
The dose metrics of concern (Table 2) were estimated continuously, as well as at specific ages (1, 3, and 6 months and 1, 5, 10, 15, 25, 50, and 75 years). It is important to note that the age-dependent values of the oral dose metrics obtained with these simulations only provide a basis for comparing the relative internal exposure associated with the same nominal daily intake and do not reflect age-dependent differences in intake. In particular, no attempt was made to evaluate life-stage-dependent exposure, such as breast- versus bottle-fed infants. This age-independent daily intake scenario was selected to provide information on the nature of the variation in the internal, or biologically effective, dose associated with exposure at different life stages to a specified acceptable daily intake (ADI) or reference dose (RfD).

For each of the surrogate chemicals, partition coefficients from the literature or from a previously published adult human model for that chemical were used (Table 2); adult metabolic and urinary clearance parameters were also based on the previously developed models. However, except as already described for the case of isopropanol, the resulting age-dependent model was not directly validated against the corresponding published adult model or against data for the surrogate chemicals.

Isopropanol.
Parameters for isopropanol were taken from the model of Clewell et al. (2001a)Go. Age-dependent metabolism of isopropanol was based on data for ADH, while age-dependent metabolism of acetone was based on data for CYP2E1. In the case of isopropanol, predictions of age-dependent dosimetry were conducted for three routes of exposure: oral, dermal, and inhalation. Artificial continuous exposure scenarios for dermal exposure (0.07 mg/l over 18.5% body surface area for males and 20% body surface area for females) and inhalation exposure (1 ppb continuous) were selected to result in values of the dose metrics on the same order of magnitude as the oral exposure, so comparisons could readily be made of the impact of route of exposure on age-dependent behavior. However, while the oral exposure was characterized as a constant daily intake, as discussed above, the inhalation and dermal exposures were characterized as a constant media concentration (in the air or in a water vehicle on a constant fraction of the total skin surface area). Thus the dependence of the exposure scenarios on body-weight, and hence age, varies significantly across routes. A more complete evaluation of age-dependent pharmacokinetics from inhalation exposures was performed in a separate study (Sarangapani et al., 2003Go).

Vinyl chloride.
The parameters for vinyl chloride were taken from a published model (Clewell et al., 2001aGo), and the age-dependence of metabolism was based on data for CYP2E1.

Methylene chloride.
The parameters for methylene chloride were taken from the model of Andersen et al. (1987)Go. The age-dependence of oxidative metabolism was based on data for CYP2E1, while the age-dependence of glutathione conjugation was based on data for ADH.

Perchloroethylene.
The parameters for perchloroethylene were taken from the model of Gearhart et al. (1993)Go, and the age-dependence of metabolism was based on data for CYP2E1. The Gearhart et al. (1993)Go model used a single-compartment model for the metabolite. Therefore, the tissue–blood partitions for the metabolite in the life-stage model were set to a uniform value producing the same volume of distribution as the Gearhart et al. (1993)Go model. Similarly, the Gearhart et al. (1993)Go model described the amount of TCA produced as 60% of the total metabolized amount of perchloroethylene; therefore, the life-stage model was parameterized such that 60% of perchloroethylene is metabolized to TCA. The adult urinary clearance value for TCA was also taken from Gearhart et al. (1993)Go, and the age-dependent urinary clearance for TCA was modeled as a fraction of the adult urinary clearance value, where the fraction was determined by the age-dependent renal blood flow data, assuming active secretion.

Nicotine.
The parameters for nicotine were taken from the model of Robinson et al. (1992)Go. The metabolism of nicotine is mainly via CYP2A6, but due to the lack of age-related data on CYP2A6, metabolism was modeled using available data on age-specific activity of CYP2C (Treluyer et al., 1997Go). The same was true for the metabolite, cotinine, which is also metabolized via the CYP2A6 pathway. Because cotinine accounts for 80% of the metabolized amount of nicotine (Robinson et al., 1992Go), the model was parameterized such that 80% of nicotine was metabolized to cotinine. Urinary clearance of both nicotine and cotinine was modeled as a fraction of the adult urinary clearance value from Robinson et al. (1992)Go, where the fraction was determined by the equations for GFR.

TCDD.
Partition coefficients for TCDD were obtained from Murphy et al. (1995)Go, and the adult metabolism parameter was adjusted to obtain a half-life of 7.5 years for a young adult, after Andersen et al. (1997)Go. Age-dependent metabolism was modeled using data for CYP1A2. Time-series sensitivity coefficients were also calculated for TCDD to demonstrate the dynamic age-dependent impact of the most sensitive parameters, the fat volume and the metabolic rate constant.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUPPLEMENTARY DATA
 REFERENCES
 
In general, the simulations of the different surrogate chemicals with the life-stage model did not produce significant differences in dose metrics between males and females; therefore, only the male results are presented for most of the chemicals. For those cases for which a significant difference was observed, both the male and female results are presented. A cross-chemical comparison of the average internal dose metrics predicted during each of the selected life stages, normalized to the 25-year-old adult value, is provided in Table 3. The entries under each life stage represent the relative sensitivity, from a pharmacokinetic viewpoint, for exposure during that life stage as compared to exposure of a 25-year-old adult. The entries for overall HKAF for each chemical represent the highest ratio of the internal dose metric for any of the life stages to the internal dose metric for a 25-year-old adult. These results can be compared with the default value for HKAF of 3.2. In general, the greatest departures from the 25-year-old adult internal dose metrics tended to occur during the infant life stage (birth to 6 months). Table 4 provides a similar comparison for one of the chemicals, isopropanol, across all three routes of exposure. For isopropanol exposure, the predicted internal dose metrics during the infant life stage are greater than the adult values for all exposure routes, with inhalation showing the greatest difference.


View this table:
[in this window]
[in a new window]
 
TABLE 3 Ratio of Average Daily Dose During Different Life stages to Average Daily Dose for (25-Year Old) Adult

 

View this table:
[in this window]
[in a new window]
 
TABLE 4 Ratio of Average Daily Dose During Different Life Stages to Average Daily Dose for (25-Year Old) Adult Resulting from Exposure to Isopropanol Via Different Routes

 
In the case of chronic toxicity, the contribution of each life stage to the cumulative internal dose over a lifetime depends on the value of the internal dose metric and the duration of the life stage. Table 5 provides the predicted contributions of each life stage to the lifetime cumulative internal dose for each chemical. Although the infant life stage demonstrated the greatest departures from the 25-year-old adult, the short duration of this life stage limits its contribution to a small percentage of the lifetime cumulative internal dose. Of course, this comparison only reflects predicted pharmacokinetic differences. Age-dependent pharmacodynamic differences should also be considered in determining the relative contribution of exposure during a particular life stage to cumulative lifetime risk for a particular effect.


View this table:
[in this window]
[in a new window]
 
TABLE 5 Fraction of Total Cumulative Lifetime Dose Contributed by Cumulative Doses During Individual Life Stages

 
Oral Exposure to Volatiles
The results of the model simulations for isopropanol (Fig. 2A) indicate that, for a constant daily intake, blood concentrations in the first decade of life decrease by nearly a factor of two, after which a slight rise is observed over the next decade. A slight decline in blood concentration is again predicted from age 20 until approximately 40, after which the blood concentration remains fairly constant. Changes in blood concentrations of the metabolite acetone parallel the initial decline for isopropanol, but return completely to perinatal levels by age 20, after which they remain relatively constant before rising slightly late in life. Predicted blood concentrations of isopropanol and acetone vary approximately 2.5- and 2-fold, respectively, across all life stages, with peak concentrations of both occurring in early life; a second peak for acetone occurs between the ages of 15 and 20 years. The more rapid fluctuations in blood concentrations superimposed over these trends reflect transient changes in hepatic clearance, principally the result of changes in liver clearance associated with growth spurts for that organ. The open circles and squares in Figure 2A represent the concentrations predicted with the published adult model (Clewell et al., 2001bGo), using the appropriate physiological and biochemical parameters for several discrete ages, as a check on the predictions of the continuous age-dependent simulations performed with the life-stage model.

The profile over time predicted for vinyl chloride arterial blood concentrations (Fig. 2B) is similar to that for isopropanol. Predicted blood concentrations decrease during the first 5 years of life, rising slightly thereafter until the age of 16, after which they decrease slightly before plateauing. Estimated vinyl chloride concentrations vary roughly three-fold over the lifetime, with the highest concentrations occurring in the first month of life. The estimated rate of reactive metabolite production per volume of liver, on the other hand, rises rapidly from birth until about age 16, after which it remains relatively constant before rising again late in life. The rate of metabolite production per volume of liver, which is the dose metric used in the cancer risk assessment for vinyl chloride (Clewell et al., 2001aGo) varies four-fold from birth to 75 years of age, with peak values estimated in adolescence at age 14 to 16 and again at the end of the simulation.

The temporal profile for methylene chloride arterial blood concentrations (Fig. 2C) is very similar to that for vinyl chloride. Methylene chloride concentrations vary three-fold over the lifetime, with the highest concentrations occurring in the first month of life. At the low concentration used in these simulations, the first-order metabolic clearance of methylene chloride by GSH conjugation competes with a higher affinity clearance by CYP2E1. The GSH conjugation rate per volume of liver, which is the dose metric used in the cancer risk assessment for methylene chloride (Andersen et al., 1987Go), varies thirty-one-fold with age, increasing consistently until age 25, after which it remains stable with only a slight decline after age 60.

In contrast with the results for isopropanol, vinyl chloride, and methylene chloride, blood concentrations of perchloroethylene were predicted to rise consistently from birth to approximately age 40, followed by a plateau between ages 40 and 50, and a decline from age 68 to 75 (Fig. 2D). Similar trends are predicted for both males and females, but female blood concentrations are predicted to be as much as 30% higher than male values. Female blood concentrations rise more rapidly between the ages of 10 and 20, a period during which females typically experience larger increases in the fractional volume of fat, an important storage compartment for the highly lipophilic perchloroethylene, as compared with males. Across the lifespan of both males and females, perchloroethylene blood concentrations were predicted to vary approximately five to seven-fold, with peak values occurring between the ages of 50 and 70.

The general rise in perchloroethylene blood concentrations, as contrasted with the relatively flat profile for the other volatiles, can be ascribed two factors: (1) the much lower metabolic and pulmonary clearance of perchloroethylene relative to the other volatiles, and (2) its much higher lipophilicity. These characteristics of perchloroethylene result in greater storage of perchloroethylene in fat and other tissues.

TCA metabolite concentrations in males and females generally parallel blood perchloroethylene concentrations (Fig. 2D); however, the range in metabolite blood concentrations across age is significantly larger than that of the parent compound.

Oral Exposure to Nonvolatiles or Semivolatiles
After a substantial decline immediately following birth, TCDD concentrations were estimated to return to perinatal levels during the first 7 years of life, followed by a second rise during adolescence, a decline until age 50, and a final rise late in life (Fig. 3A). At birth the concentration of TCDD in the neonate reflects transplacental exposure to maternal stores of the chemical. The initial drop in TCDD concentration during the neonatal period results from dilution of TCDD stores by the rapid growth of the neonate. This prediction of the model, while somewhat nonintuitive, is consistent with experimental data showing that TCDD concentrations in neonates are below adult levels (Kreuzer et al., 1997Go). As pointed out by Kreutzer et al. (1997)Go, the rapid growth during the neonatal period, and the resulting dilution of TCDD stores, produces an apparent half-life for TCDD in infants on the order of 5 months, as opposed to a half-life observed in the adult (resulting primarily from metabolic clearance) on the order of 10 years.



View larger version (17K):
[in this window]
[in a new window]
 
FIG. 3. Blood concentrations as a function of age for (A) TCDD for continuous oral exposure of males (solid line) or females (dashed line) at a constant daily intake of 1 ng/kg/d, and (B) nicotine and its metabolite cotinine for continuous oral exposure to nicotine at a constant daily intake of 1 µg/kg/d.

 
The rise in TCDD blood concentrations during childhood and adolescence can be attributed to the accumulation of the highly lipophilic, poorly metabolized compound in fat as the growth rate declines and the apparent half-life decreases. Male and female blood concentrations diverge somewhat after the 7-year point, because the fractional volume of fat in females rises more rapidly than in males and rises to values that are about 50% larger than those for males (Supplementary Data, Table B). The larger storage capacity for TCDD in females results in blood concentrations that are generally lower than those for males.

Time-dependent sensitivity coefficients demonstrate the dependence of blood concentrations on metabolic clearance and the size of the fat compartment (Fig. 4). Blood TCDD concentrations are more sensitive to fat compartment size than clearance during childhood and adolescence, while the rate of clearance exerts more control over blood concentrations during adulthood. This result demonstrates that, where a large number of parameters change with age, the relative importance of the parameters can also change with age, resulting in different expectations regarding risk factors in different age groups. The investigation of these age-dependent sensitivities would be impossible with the standard ''reference man,'' young-adult PBPK models typically used in risk assessment.



View larger version (13K):
[in this window]
[in a new window]
 
FIG. 4. Age-dependent sensitivity coefficients for blood TCDD concentrations resulting from constant oral exposure. Arterial blood concentration is more sensitive to changes in fat volume (solid line) than metabolism for the first several years of life, after which it becomes more sensitive to the metabolic rate constant (dotted line). Other parameters showed lower sensitivity.

 
In contrast to the behavior of the highly lipophilic TCDD, estimated blood concentrations for the water-soluble compound nicotine decline steadily from birth until approximately age 20, when the concentrations plateau (Fig. 3B). This trend correlates with increases in total hepatic metabolic capacity, with the spikes in blood concentration reflecting changes in liver blood flow. With the exception of an initial rise in concentration during early life driven by an increase in metabolic capacity in the liver, the trend for the cotinine metabolite blood concentration is similar to the trend for nicotine. Blood concentrations of both the parent and metabolite rise slightly after 60 years of age, as metabolic and renal clearances diminish.

Cross-Route Comparison
Comparison of dose metrics for isopropanol across oral, inhalation, and dermal exposures confirms the expectation that the age-dependent pattern of internal exposure is different for each of the exposure routes (Table 4). For inhalation (Fig. 5A), higher concentrations of both parent and metabolite were predicted to occur during early life. This higher exposure during early life was much more pronounced for inhalation than for ingestion (Fig. 2a). For the dermal route (Fig. 5B), on the other hand, peak concentrations of parent were predicted to occur late in life, although peak concentrations of metabolite were still predicted to occur early in life. The variation in IPA blood concentrations across age was less that two-fold, regardless of route of exposure, but the variation in acetone blood concentrations was much greater, ranging from two- to eight-fold. The open circles and squares in Figure 5A represent the concentrations predicted with the published adult model (Clewell et al., 2001bGo), using the appropriate physiological and biochemical parameters for several discrete ages as a check on the predictions of the time-dependent life-stage model.

The comparison of the age-dependent concentrations for isopropanol and acetone across exposure routes provides a classic demonstration of the first pass effect of presystemic hepatic metabolism. Following oral exposure, all of the administered isopropanol goes to the liver before distribution, resulting in the generation of greater blood concentrations of acetone, compared with inhalation or dermal exposure.


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUPPLEMENTARY DATA
 REFERENCES
 
The results of the simulations conducted with the life-stage PBPK model (Tables 3 and 4) indicate that, in general, variations in pharmacokinetic dose metrics for a chemical, averaged over different life stages, were within a factor of 2 of the young adult values. As discussed in the results section, larger variations were observed, particularly for exposures very early in life, but these were associated with relatively short durations. For all but one of the surrogate chemicals studied, estimated variation in the dose metrics over the lifespan of an individual following drinking water exposure is less than the IPCS (2001)Go default factor for human pharmacokinetic variability of 3.2 (Table 3). For nicotine, the estimated average dose metric for the parent chemical during the birth to 6-month life stage is approximately 3.4 times greater than the parent concentration estimated at 25 years of age.

Similar trends in parent chemical dose metrics were observed for all of the volatiles, with the exception of perchloroethylene (Table 3). For most of the volatile chemicals, the average dose metrics peaked early in life, with decreasing trends over the lifespan. For these parent chemicals, the adjustment factor for pharmacokinetic variability (HKAF) is based on the ratio for the neonatal (birth to 6 months) life stage. For perchloroethylene, however, accumulation of the parent chemical over time results in increased concentrations later in life, rather than early. A similar trend was also observed with TCDD.

Impact of Exposure Scenario
The direction and magnitude of the age-related changes in isopropanol arterial blood concentration as a function of route of exposure can best be understood in terms of age-related changes in the ratio of (dose rate)/(clearance) resulting from differential body-weight, and hence age-dependent, scaling of these terms. For each exposure route, dose rate is a different function of body weight, and therefore of age. In these case studies, the dose rate for oral exposure is assumed to be a zero-order intake scaled by body weight. That is, the exposure comparison across ages is made on the basis of equal intake of the chemical in mg/kg/d, rather than equal media (e.g., drinking water) concentration. Thus, the oral dose rate scales linearly with body weight. An alternative comparison could be made assuming equal drinking water concentration and adding age-specific drinking water consumption data to the model. Since infants and children consume up to twice as much drinking water per kg body weight as adults (USEPA, 1997Go), this assumption would tend to increase the dose metrics predicted for these earlier life stages by a factor of up to 2 for oral exposure.

The inhalation exposures in these case studies were conducted at a constant inhalation concentration. For inhalation exposure to isopropanol, which is highly soluble in the blood, the dose rate is approximately equal to the exposure concentration multiplied by alveolar ventilation rate (Supplementary Data, Appendix B). Alveolar ventilation rate scales less than linearly with body weight, in part because of the nature of the scaling of basal oxygen consumption and in part because of the higher activity level in children relative to adults. Therefore, the inhalation dose rate for isopropanol is a somewhat smaller function of body weight than the oral dose rate. After an initial loading period, the uptake from inhalation of less soluble chemicals is roughly independent of ventilation rate, depending instead primarily on metabolic clearance. For these chemicals, blood concentration is primarily determined by the blood—air partition coefficient, which is assumed in these case studies to be independent of age. Thus "inhaled dose" is not a well-defined concept for poorly soluble, highly volatile chemicals such as vinyl chloride.

For dermal exposure, the dose rate is the exposure concentration x surface area x permeability coefficient. In these case studies, it has been assumed that (1) exposure is to a constant concentration, (2) exposed skin area is a constant fraction of the total body skin area, and (3) there is no change in the permeability of the skin to the chemicals with age. Thus the dose rate for dermal exposure in these case studies scales with surface area, which is a smaller function of body weight than ventilation rate. If, instead, the dermal comparison were made on the basis of equal dermal uptake in mg/kg/day, the concentrations in the infant and child would be increased by a factor of up to about 2.

Clearance, whether metabolism or blood-flow limited, tends to scale roughly with surface area over most of the lifetime, regardless of the route of exposure. However, in the first decade of life, the development of metabolic and excretory systems can lead to a substantial increase in clearance. The combination of these dose-rate and clearance factors leads to the complex differences in the age-dependent profiles, such as those shown in Figures 2A, 5A, and 5B for the concentration of isopropanol in arterial blood for the different exposure routes. An analysis of the factors contributing to these steady-state behaviors is provided in the supplementary data, Appendix B.

Additional Considerations and Uncertainties
It is important to note that the purpose of this investigation was to develop a methodology for using PBPK modeling to evaluate the potential impact of age- and gender-specific differences on risk from chemical exposure, and to demonstrate that methodology with representative chemicals. It is not the intent of the authors to suggest that the current predictions of this initial model for the specific chemicals addressed in the case studies should be used quantitatively in support of risk assessments. A number of simplifying assumptions have been necessary to perform these case studies in the face of significant uncertainties regarding the age-dependence of the clearance of the chemicals concerned. The main criterion for the selection of the surrogate chemicals was the existence of a validated human PBPK model in the published literature. The available human PBPK models, however, were strictly for adults, and the available age-specific metabolic information was extremely limited.

For example, for chemicals metabolized by the CYP enzymes, we have assumed that a single isoform is solely responsible for metabolic clearance regardless of age. However, other CYP isoforms can contribute to the metabolism of these chemicals, and the age-dependence of these other isoforms could vary considerably from the isoform selected for a particular case study. Many chemicals, like nicotine, have complex metabolic profiles in which minor pathways could play an important role in specific tissues at certain ages. In conducting an age-dependent risk assessment for a specific chemical, these possibilities would have to be explored.

In the case of nicotine it was even necessary to use data on the ontogeny of CYP2C as a surrogate for the development of CYP2A6. This was also a limitation for evaluating age- and gender-related changes in gluthathione-S-transferase. Very little is known about the development of this enzyme, although the limited data (Mendrala et al., 1993Go; Pacifici et al., 1981Go) suggests a pattern similar to that for ADH (Pikkarainen and Räihä, 1967Go). Therefore, it was necessary to use the quantitative information on ADH to characterize the development of GST.

Assumptions were also made regarding the stoichiometry of the metabolism of the parent compound. For both PERC and nicotine, the percentage of parent chemical that is metabolized to the major metabolite (TCA and cotinine, respectively) was expressed as a constant fraction across the lifespan, due to a lack of age-specific information to the contrary. However, this yield may not be constant throughout the lifespan. Another metabolic issue that was not addressed in these case studies was detoxification of reactive metabolites (e.g., the epoxide formed from vinyl chloride). For some chemical intermediates, the rate of clearance may also be subject to age-specific variation.

There is even greater uncertainty regarding the nature of clearance processes for highly lipophilic chemicals such as TCDD. The clearance of TCDD has been ascribed to metabolism and biliary/fecal excretion (Rohde et al., 1999Go). For this case study, we have assumed that the age-dependence of the clearance of TCDD is primarily dependent on the development of metabolic clearance by CYP1A2. This assumption was based on data showing that fecal clearance of nonmetabolized TCDD contributes on the order of 37% to total elimination in the adult (Rohde et al., 1999Go). To the extent that fecal clearance of nonmetabolized chemical contributes to the age-dependent clearance of TCDD, the results of this case study could be somewhat misleading. However, the most striking feature of the age-dependent kinetics of TCDD, the rapid decrease in body burden during the first year of life, is not sensitive to this uncertainty. Regardless of the nature of TCDD clearance, the observed kinetics in infants is consistent with complete absorption and negligible elimination during the first year of life (Abraham et al., 1996Go), with an apparent half-life on the order of 5 months (as compared to greater than 5 years in the adult) that reflects dilution by the growth of tissues rather than actual elimination (Kreuzer et al., 1997Go).

Conclusions
The results of the present analysis for environmental contaminants are in general agreement with the findings of quantitative analyses of data on pharmaceutical chemicals (Ginsberg et al., 2002Go; Hattis et al., 2003Go; Renwick et al., 2000Go), which have suggested that the largest difference in pharmacokinetics observed between children and adults is for the early postnatal period. These results are also consistent with work conducted by Alcorn and McNamara (2003)Go in which the general pattern of postnatal development of selected CYP P450 enzyme pathways were quantified based on in vitro activity in fetal or infant hepatic microsomes as a fraction of adult activity. The differences observed in the pharmacokinetics between children and adults, which are reflected in a slower clearance of the chemicals in the infant, appear to be related primarily to the immaturity of the metabolic enzyme systems responsible for clearance of these chemicals from the body. However, these enzyme systems mature rapidly, resulting in smaller differences in pharmacokinetics as compared to adults (generally on the order of a factor of 2 or 3) by the time children reach 6 months of age. Changes observed in later childhood, which can result in faster clearance than in the adult, appear to be related to the physiological changes associated with growth (i.e., changes in organ-to-body-weight ratios and accompanying changes in organ blood flows) and activity (which impacts the ventilation rate).

For selected classes of chemicals, mainly the more lipophilic compounds (perchloroethylene and TCDD), there is a relative increase in dose metrics in the later life stages (25 years to 75 years) for either the circulating metabolite or the parent compound. This increase reflects the capability of these chemicals, due to their physical/chemical properties, to accumulate in the body over the duration of the lifetime, possibly combined later in life with a decrease in function of clearance systems.

The results of the current analyses provide some insight into potential differences in pharmacokinetics across life stages and how these differences may result in differences in internal dose metrics following chemical exposure. The most important factor appears to be the potential for decreased clearance of a toxic chemical in the perinatal period due to the lower activity of many metabolic enzyme systems, although this same factor may also reduce the production of a toxic metabolite. In general, as long as large differences in dosimetry are restricted to the early childhood period, their contribution to the lifetime average daily dose, the metric used for cancer risk assessment and many chronic noncancer effects, is limited by the relatively small portion of the lifetime over which they occur (Table 5). However, due to the potential for age-dependent pharmacodynamic factors associated with growth and development, there may be chemicals and health outcomes for which the maintenance of a higher internal dose over a relatively brief period during early life may have a substantial impact on risk (Ginsberg, 2003Go). Thus, the immaturity of clearance systems during early life may represent an important window of pharmacokinetic sensitivity for exposures to toxic chemicals.

The results presented here illustrate a methodology for using PBPK modeling to evaluate the potential impact of age- and gender-specific differences on risk from chemical exposure. This modeling approach establishes a basis for forming expectations about the magnitude and range of differences in dosimetry resulting from age-dependent differences in various biochemical and physiological parameters at different life stages. This work represents an initial attempt to provide a predictive pharmacokinetic framework that could be used to characterize the effect of age and gender differences on tissue dosimetry for a chemical or class of chemicals. In particular, only young adult data has been used for validation of chemical-specific model predictions; therefore, extrapolations with the model to other life stages, while based on a reasonable description of age-dependent physiological and biochemical processes, have not yet been validated. At this stage, the model predictions for early and late life stages should be considered to represent reasonable expectations rather than predictive extrapolations.


    SUPPLEMENTARY DATA
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUPPLEMENTARY DATA
 REFERENCES
 
Supplementary data for this paper include the data and equations describing growth- and age-related parameters used in model development, and steady state analyses, of the factors determining age-dependent behavior.


    ACKNOWLEDGMENTS
 
This project was funded by a grant from the American Chemistry Council, but the conclusions are those of the authors. The authors wish to thank Rebecca Clewell, CIIT Centers for Health Research, Research Triangle Park, NC, for her helpful review of a draft of this manuscript.


    NOTES
 
2 Current address: Novartis Pharmaceuticals, East Hanover, NJ 07936. Back

1 To whom correspondence should be addressed at ENVIRON Health Sciences Institute, 602 East Georgia Avenue, Ruston, LA 71270


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUPPLEMENTARY DATA
 REFERENCES
 
Abraham, K., Knoll, A., Ende, M., Papke, O., and Helge, H. (1996). Intake, fecal excretion, and body burden of polychlorinated dibenzo-p-dioxins and dibenzofurans in breast-fed and formula-fed infants. Pediatr. Res. 40(5), 671–679.[Abstract]

Alcorn, J., and McNamara, P. J. (2003). Pharmacokinetics in the newborn. Adv. Drug Deliv. Rev. 55, 667–686.[CrossRef][ISI][Medline]

Andersen, M. E., Gargas, M. L., Hays, S. M., Clewell, H. J., III, and Paustenbach, D. J. (1997). Estimating minimally effective inducing doses (ED01s) of TCDD in livers of rats and humans using physiologically based pharmacokinetic modeling. Organohalogen Compounds 34, 311–316.

Andersen, M., Clewell, H., Gargas, M., Smith, F., and Reitz, R. (1987). Physiologically based pharmacokinetics and the risk assessment process for methylene chloride. Toxicol. Appl. Pharmacol. 87(2), 185–205.[ISI][Medline]

Åstrand, I. (1983). Effect of physical exercise on uptake, distribution and elimination of vapors in man. In Modeling of Inhalation Exposure to Vapors: Uptake, Distribution, and Elimination (Fiserova-Bergerova V, Ed.), Vol. 2, pp. 107–130. CRC Press, Boca Raton, FL.

Braunlich, H. (1977). Kidney development: Drug elimination mechanisms. In Drug Disposition During Development (P. Morselli, Ed.), pp.89–101. Spectrum Publications Inc., New York.)

Clewell, H. J., Gentry, P. R., Gearhart, J. M., Allen, B. C., and Andersen, M. E. (2001a). Comparison of cancer risk estimates for vinyl chloride using animal and human data with a PBPK model. Science of the Total Environment 274(1–3), 37–66.[CrossRef][ISI][Medline]

Clewell, H. J., Gentry, P. R., Gearhart, J. M., Covington, T. R., Banton, M. I., and Andersen, M. E. (2001b). Development of a physiologically based pharmacokinetic model of isopropanol and its metabolite acetone. Toxicol. Sci. 63, 160–172.[Abstract/Free Full Text]

Clewell, H., Teeguarden, J., McDonald, T., Sarangapani, R., Lawrence, G., Covington, T., Gentry, R., and Shipp, A. (2002). Review and evaluation of the potential impact of age and gender-specific pharmacokinetic differences on tissue dosimetry. Crit. Rev. Toxicol. 32(5), 329–389.[ISI][Medline]

Gearhart, J. M., Mahle, D. A., Greene, R. J., Seckel, C. S., Flemming, C. D., Fisher, J. W., and Clewell, H. J. (1993). Variability of physiologically-based pharmacokinetic (PB-PK) model parameters and their effects on PB-PK model predictions in a risk assessment for perchloroethylene. Toxicol. Lett. 68, 131–144.[CrossRef][ISI][Medline]

Gentry, P. R., Covington, T. R., Andersen, M. E., and Clewell, H. J. (2002). Application of a physiologically-based pharmacokinetic model for isopropanol n the derivation of an RfD/RfC. Reg. Toxicol. Pharmacol.. 36, 51–68.[CrossRef][ISI][Medline]

Gentry, P. R., Covington, T. R., and Clewell, H. J. (2003). Evaluation of the potential impact of pharmacokinetic differences on tissue dosimetry in offspring during pregnancy and lactation. Reg. Toxicol. Pharmacol. 38(1), 1–16.[CrossRef][ISI][Medline]

Ginsberg, G. L. (2003). Assessing cancer risks from short-term exposures in children. Risk Anal. 23(1), 19–34.[CrossRef][ISI][Medline]

Ginsberg, G., Hattis, D., Sonawane, B., Russ, A., Banati, P., Kozlak, M., Smolenski, S., and Goble, R. (2002). Evaluation of child/adult pharmacokinetic differences from a database derived from the therapeutic drug literature. Toxicol. Sci. 66(2), 185–200.[Abstract/Free Full Text]

Guengerich, F., Kim, D., and Iwasaki, M. (1991). Role of cytochrome P-450 IIE1 in the oxidation of many low molecular weight cancer suspects. Chem. Res. Toxicol. 4, 168–179.[ISI][Medline]

Hattis, D., Ginsberg, G., Sonawane, B., Smolenski, S., Russ, A., Kozlak, M., and Goble, R. (2003). Differences in pharmacokinetics between children and adults—II. Children's variability in drug elimination half-lives and in some parameters needed for physiologically-based pharmacokinetic modeling. Risk Anal. 23(1), 117–142.[CrossRef][ISI][Medline]

International Commission on Radiological Protection (ICRP). (1975). Report of the Task Group on Reference Man. Pergamon Press, Oxford. ICRP Publication 23.

International Programme on Chemical Safety (IPCS). (2001). Guidance document for the use of data in development of chemical-specific adjustment factors (CSAF) for interspecies differences and human variability in dose/concentration response assessment World Health Organization, Geneva. Available at: http://www.ipcsharmonize.org.

Kreuzer, P. E., Csanady, G. A., Baur, C., Kessler, W., Paepke, O., Greim, H., and Filser, J. G. (1997). 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) and congeners in infants. A toxicokinetic model of human lifetime body burden by TCDD with special emphasis on its uptake by nutrition. Arch. Toxicol. 71, 383–400.[CrossRef][ISI][Medline]

Lipscomb, J., Meek, M. E., Krishnan, K., Kedderis, G. L., Clewell, H. J., and Haber, L. (2004). Incorporation of pharmacokinetic and pharmacodynamic data into risk assessments. Toxicology Mechanisms and Methods, 14, 1–14.[ISI]

Mayersohn, M. (1994). Pharmacokinetics in the elderly. Environ. Health Perspect. 102, 119–124.

Mendrala, A. L., Langvardt, P. W., Nitschke, K. D., Quast, J. F., and Nolan, R. J. (1993). In vitro kinetics of styrene and styrene oxide metabolism in rat, mouse, and human. Arch. Toxicol. 67, 18–27.[ISI][Medline]

Milsap, R. L., and Jusko, W. J. (1994). Pharmacokinetics in the infant. Environ. Health Perspect. 102 (Suppl. 11), 107–110.[ISI][Medline]

Murphy, J., Janszen, D., and Gargas, M. (1995). An in vitro method for determination of tissue partition coefficients of non-volatile chemicals such as 2,3,7,8-tetrachlorodibenzo-p-dioxin and estradiol. J. Appl. Toxicol. 15(2), 147–152.[ISI][Medline]

National Research Council (NRC). (1986). Dose-route extrapolations: Using inhalation toxicity data to set drinking water limits. In Drinking Water and Health. Vol. 6, Chap. 6. National Academy Press, Washington, DC.

National Health and Nutrition Examination Survey (NHANES). (1995). Third National Health and Nutrition Examination Survey, 1988–1991. Selected Laboratory and Mobile Examination Center Data. Version 1, September 1995.

Ogiu, N., Nakamura, Y., Ijiri, I., Hiraiwa, K., and Ogiu, T. (1997). A statistical analysis of the internal organ weights of normal Japanese people. Health Phys. 72, 368–383.[ISI][Medline]

Pacifici, G. M., Boobis, A. R., Brodie, M. J., McManus, M. E., and Davies, D. S. (1981). Tissue and species differences in enzymes of epoxide metabolism. Xenobiotica 11, 73–79.[ISI][Medline]

Pelekis, M., Gephart, L., and Lerman, S. (2001). Physiological-model-based derivation of the adult and child pharmacokinetic intraspecies uncertainty factors for volatile organic compounds. Reg. Toxicol. Pharmacol. 33, 12–20.[CrossRef][ISI][Medline]

Pikkarainen, P. H., and Räihä, N. C. R. (1967). Development of alcohol dehydrogenase activity in the human liver. Pediatr. Res. 1, 165–168.[ISI][Medline]

Plunkett, L. M., Turnbull, D., Rodricks, J. V. (1992). Differences between adults and children affecting exposure assessment. In Similarities and Differences Between Children and Adults: Implications for Risk Assessment (P. Guzelian, C. Henry, and S. Olin, Eds.), pp. 79–94. ILSI Press, Washington, DC.

Renwick, A., Dorne, J., and Walton, K. (2000). An analysis of the need for an additional uncertainty factor for infants and children. Reg. Toxicol. Pharmacol. 31, 286–296.[CrossRef][ISI][Medline]

Robinson, D. E., Balter, N. J., and Schwartz, S. L. (1992). A physiologically based pharmacokinetic model for nicotine and cotinine in man. J. Pharmacokinet. Biopharm. 20, 591–609.[ISI][Medline]

Rohde, S., Moser, G. A., Papke, O., and McLachlan, M. S. (1999). Clearance of PCDD/Fs via the gastrointestinal tract in occupationally exposed persons. Chemosphere 38(14), 3397–3410.[CrossRef][ISI][Medline]

Sarangapani, R., Gentry, P. R., Covington, T. R., Teeguarden, J. G., and Clewell, H. J. (2003). Evaluation of the potential impact of age- and gender-specific lung morphology and ventilation rate on the dosimetry of vapors. Inhal. Toxicol. 15(10), 987–1016.[ISI][Medline]

Sonnier, M., and Cresteil, T. (1998). Delayed ontogenesis of CYP1A2 in the human liver. Eur. J. Biochem. 251, 893–898.[Abstract]

Treluyer, J. M., Gueret, G., Cheron, G., Sonnier, M., and Cresteil, T. (1997). Developmental expression of CYP2C and CYP2C-dependent activities in the human liver: In-vivo/in-vitro correlation and inducibility. Pharmacogenetics 7, 441–452.[ISI][Medline]

U. S. Environmental Protection Agency (USEPA). (1997). Exposure Factors Handbook Volume I: General Factors. Office of Research and Development, Washington, D. C., EPA/600/P-95/002FA.

Vieira, I., Sonnier, M., and Cresteil, T. (1996). Developmental expression of CYP2E1 in the human liver. Hypermethylation control of gene expression during the neonatal period. Eur. J. Biochem. 238, 476–483.[Abstract]





This Article
Abstract
FREE Full Text (PDF)
Supplementary Data
All Versions of this Article:
79/2/381    most recent
kfh109v1
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (6)
Disclaimer
Request Permissions
Google Scholar
Articles by Clewell, H. J.
Articles by Teeguarden, J. G.
PubMed
PubMed Citation
Articles by Clewell, H. J.
Articles by Teeguarden, J. G.