ToxSci Advance Access originally published online on August 22, 2008
Toxicological Sciences 2008 106(2):519-537; doi:10.1093/toxsci/kfn176
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Estimates of Cancer Potency of 2,3,4,7,8-Pentachlorodibenzofuran Using Both Nonlinear and Linear Approaches
* Ted Simon, LLC, Winston, Georgia 30187
The Sapphire Group, Inc., Cleveland, Ohio 44122
Summit Toxicology, LLP, Falls Church, Virginia 22044
Dow Chemical Company, Midland, Michigan 48674
1 To whom correspondence should be addressed at Ted Simon, LLC, 4184 Johnston Road, Winston, GA 30187. Fax: 770-942-3424. E-mail: ted{at}tedsimon-toxicology.com.
Received May 14, 2008; accepted August 14, 2008
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Cancer potency estimates were derived for 2,3,4,7,8-pentachlorodibenzofuran (4-PeCDF) using data collected from the recently published National Toxicology Program bioassay in female Sprague-Dawley rats. By using a toxicokinetic model for 4-PeCDF, the dose-response relationship for combined liver tumors (hepatocellular adenomas and cholangiocarcinomas) in rats was assessed in terms of lifetime average liver concentration and lifetime average adipose concentration with data from both the lifetime and the stop-exposure components of the bioassay. Benchmark dose modeling was performed to estimate tissue concentrations at two points of departure (EC10 and EC01 and their 95% upper and lower confidence limits). The same toxicokinetic model with human input values was then used to back-extrapolate human equivalent doses that corresponded to the internal tissue concentration measures at the points of departure. Information regarding the cancer mode of action was used to support the development of several toxicity criterion values based on a nonlinear method, e.g., reference dose or tolerable daily intake. Nonlinear estimates of toxicity criteria based on observed noncancer toxic events as possible precursors to tumor formation were also derived and were similar in value to those based on combined liver tumors. For comparison purposes, linear estimates of cancer potency were also derived.
Key Words: dioxin; 2,3,4,7,8-pentachlorodibenzofuran; dose-response; cancer potency factor; reference dose; tumor promotion.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
The risks from exposure to chlorinated dioxin and furan mixtures have traditionally been assessed as a function of their relative potency to 2,3,7,8-tetrachlorodibenzodioxin (TCDD) using a toxicity equivalence factor (TEF) approach (Haws et al., 2006
; Van den Berg et al., 2006). TEFs were originally developed in the 1980s as an "interim" method to account for the absence of congener- and isomer-specific toxicity data (Birnbaum and DeVito, 1995) and "as an interim science policy position until more definitive experimental procedures are developed for these mixtures" (USEPA, 1989).
The National Toxicology Program (NTP) recently published new bioassays for TCDD and 2,3,4,7,8-pentachlorodibenzofuran (4-PeCDF), a dioxin congener having a relatively high TEF value (NTP, 2006a,b). The cancer bioassays were conducted using a sophisticated design that incorporated six dose groups, including a stop-exposure group, and multiple measures of tissue concentrations, enzyme activity, and markers of cell proliferation over time. These cancer bioassays not only allow for the development and use of chemical-specific cancer potency factors for these compounds and enable reevaluation of the TEFs (e.g., Budinsky et al., 2006; Gray et al., 2006) but also provide an opportunity to conduct the cancer risk assessment on an internal dose basis to account for interspecies differences in toxicokinetics and mode of action (MOA) (USEPA, 2005).
We present here a cancer dose-response assessment for 4-PeCDF relying upon estimates of internal dose. We use benchmark dose (BMD) modeling and follow a nonlinear approach to derive a reference dose (RfD) or tolerable daily intake in units of dose. For comparison, we also use a linear approach to derive a traditional cancer potency factor in units of risk per dose. For an additional comparison, we use allometric scaling and traditional extrapolation factors to derive other RfD values. The exploration of both linear and nonlinear approaches to cancer risk assessment of dioxin-like compounds will enable a quantitative assessment of uncertainty associated with the MOA for carcinogenic events. This approach is the best practice, according to the United States Environmental Protection Agency's Guidelines for Carcinogen Risk Assessment (USEPA, 2005). Such an approach was recommended by the National Academy of Sciences expert panel in their recent review of USEPA's dioxin reassessment and also by other scientists (NAS, 2006; Popp et al., 2006; Walker, 2007).
Demonstrated increases in hepatocellular adenomas and cholangiocarcinomas as well as small increases in oral gingival squamous cell carcinomas occurred in the female Sprague-Dawley (SD) rats used in the bioassay (NTP, 2006a). A single cystic keratinizing epithelioma (CKE) of the lung was observed in the 200 ng/kg/day treatment group, and similar tumors showed an elevated response in rats treated with 100 ng/kg/day TCDD (NTP, 2006a,b). However, the frequency of CKE (1/53) in the rats treated with 200 ng/kg/day of 4-PeCDF was not sufficient for dose-response modeling. Historically, rodent liver tumors have been used as the basis of cancer risk assessment of dioxin-like compounds in humans (Maruyama and Aoki, 2006; USEPA, 1989, 2003). The use of this specific end point here also permits comparison with historical toxicity criteria (USEPA, 1989, 2003; WHO, 1998, 2000). According to the NTP bioassay report, the data indicated "some evidence of carcinogenic activity" (NTP, 2006a).
The occurrence of combined liver tumors, hepatocellular adenomas and cholangiocarcinomas, was the most notable carcinogenic response to 4-PeCDF observed in the bioassay. Here, two internal dose metrics are obtained from a toxicokinetic model to assess dose-response. BMD modeling at the 1% and 10% points of departure (PODs) was then used as the basis for both the nonlinear and the linear approaches to quantitative dose-response assessment. Extrapolation of the rat internal doses to humans is accomplished with the same toxicokinetic model reparameterized for humans, a methodology recommended by Clewell et al. (2002).
Lifetime average liver concentration (LALC) was chosen as the primary dose metric. At all dose levels, the concentrations of 4-PeCDF observed in the livers of the rats in the NTP bioassay were considerably higher than those observed in adipose tissue (NTP, 2006a). If the liver tissue is representative of other tissues, then this choice takes into account these high concentrations.
One of the roles of cytochrome p450 1A2 (CYP1A2) is to bind 4-PeCDF and other dioxin-like chemicals (DLCs) (Andersen et al., 1993). This binding is reversible, and there is likely equilibrium between free 4-PeCDF and that bound to CYP1A2. In addition, CYP1A2 induction appears essential to hepatic injury and eventual tumor formation due to DLCs in the liver (Tijet et al., 2006). The high liver concentrations in rats were likely due to binding to CYP1A2 and/or other proteins (Diliberto et al., 1999). Because binding of 4-PeCDF to the CYP1A2 protein is reversible, some, if not all, of the "bound" 4-PeCDF in the liver of the rats in the NTP bioassay remains potentially biologically available. As such, LALC represents the most appropriate dose metric for the end point of liver tumors.
Lifetime average adipose concentration (LAAC) was chosen as an alternate dose metric used to provide an assessment based on a marker for "free" 4-PeCDF, i.e., not bound to macromolecules.
The use of LAAC as a dose metric ignores the contribution of reversibly bound 4-PeCDF in the liver. In addition, the liver is the most susceptible tissue to the carcinogenic effects of 4-PeCDF in rodents.
In summary, liver tumors represent the most appropriate end point and LALC is the more appropriate dose metric of the two considered. We have included quantitative cancer risk estimates based on LAAC as a means of exploring potential uncertainties in these risk estimates (e.g., NAS, 2006). Finally, we also assess the quantitative impact of the choice of two PODs and the use of a more traditional linear approach on the resulting risk estimates.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Evaluation of toxicokinetic models.
The tissue concentration measurements obtained in this bioassay were evaluated with a toxicokinetic model to determine external doses in humans that would result in tissue concentrations equivalent to those observed in rats. The LALC and lifetime average adipose tissue concentrations in the SD rats used in the bioassay were estimated with the Carrier model (Aylward et al., 2004
, 2005; Carrier et al., 1995a,b). The Carrier model was also used for determination of human external doses based on tissue concentrations corresponding to the POD tissue concentrations estimated in rats (see below). Model parameters for both humans and rats are presented in Table 1. For humans, an exposure duration of 55 years, beginning at age 20 and continuing until age 75, was assumed.
|
Use of the toxicokinetic model developed and used by Maruyama et al. (2002)
was also considered. However, this model was unsuccessful in estimating tissue concentrations in rats that reflected the tissue concentrations obtained in the bioassay results (e.g., Maruyama and Aoki, 2006).
Average tissue concentrations in the rats were estimated using the Carrier model implemented on a weekly timescale. Background concentrations were used as starting points in lieu of zero in the toxicokinetic model (Ruby et al., 2004). The sum of the five daily gavage doses was used as the weekly dose. For example, the rats receiving a gavage dose of 6 ng/kg were assumed to absorb 90% of the dose. Hence, the daily absorbed dose for 5 days was 5.4 ng/kg and the weekly absorbed dose was 27 ng/kg/week.
In addition to the modeled average tissue concentrations, measured average tissue concentrations were calculated from the NTP (2006a) data using the method of Gray et al. (2006) based on the trapezoidal rule. Figures 3 and 4 show the model fits to the measured liver and adipose tissue concentrations, and Table 2 shows both the modeled tissue concentrations and those estimated from measured concentrations.
|
|
|
Benchmark Dose modeling.
Benchmark Dose (BMD) modeling was conducted using USEPA's BMD Software to obtain dose-response relationships between the poly-3 survival-adjusted incidence of combined liver tumors in rats and two internal dose measures. Two benchmark response (BMR) rates, 10% and 1%, were modeled to obtain the tissue concentrations for use as the PODs (EC10 and EC01) along with appropriate 95% confidence limits (LEC10 and LEC01; UEC10 and UEC01). The best fitting model based on the Akaike Information Criterion was chosen from the available models. Because the BMD software reports only lower confidence limits on modeled BMDs, upper confidence limits for POD values were developed by assuming that the occurrence of tumor at a given dose level was Poisson distributed and using the lower confidence limit on the Poisson rate parameter. In the Supplementary Information, additional information on goodness of fit is provided.
It should also be noted that the tissue concentrations of 4-PeCDF in the vehicle control group reported in NTP (2006a) were below the limit of quantitation (LOQ) in all fat and liver samples save the liver sample at week 105. We wished to determine if inclusion of nonzero values for the vehicle control group might change the results. Hence, in the vehicle control group, the value of 1615 pg/g was used for the liver concentration at 105 weeks and the value of 187.5 pg/g was used for the adipose tissue concentration at 105 weeks. These values are the measured liver concentration and the adipose LOQ, respectively.
The human Carrier toxicokinetic model was used to calculate external human equivalent doses (HEDs) in terms of nanogram per kilogram per day. In other words, these values were the modeled daily doses in nanogram per kilogram yielding the same LALC and LAAC dose metric values as the modeled POD tissue concentrations developed from the rat bioassay.
Choice of extrapolation factors.
Figure 1 illustrates the approach to the derivation of cancer RfDs consistent with a threshold approach to cancer dose-response. Typical extrapolation or uncertainty factor (UF) components were assessed for their relevance and appropriate values (Dorne and Renwick, 2005) and were applied on an internal or an external dose basis, as appropriate.
|
When considering data upon which the interspecies and intraspecies extrapolation factors are derived, it is important to distinguish between "sensitivity," on one hand, and "efficacy" or "responsiveness," on the other hand. Sensitivity refers to the relative dose levels at which a particular response occurs, whereas efficacy describes the magnitude of response without reference to a specific dose level. For chemicals such as 4-PeCDF, the MOA for which receptor binding is involved, sensitivity would be measured by dose-response evaluation of end points reflective of receptor binding. Responsiveness or efficacy would be measured by the greatest response, presumably occurring at the maximum dose. Dose threshold or EC50 values provide information related to sensitivity, whereas the maximal response provides information related to efficacy.
Specifically, the following UF evaluations and applications were made.
- UFa: This UF accounts for interspecies extrapolation. Because the toxicokinetic variation between species (rats and humans) is being accounted for explicitly through the use of a relevant internal dose metric for extrapolation between species, no toxicokinetic component of UFa was applied.
A recent World Health Organization (WHO) review of dioxins concluded that humans were no more sensitive and probably less sensitive than the most sensitive rodent species (WHO, 1998, 2000). Numerous studies indicate that the human aryl hydrocarbon receptor (AhR) binds TCDD and related compounds with much lower affinity than in the most sensitive rodent species due to a specific mutation in the receptor (reviewed in Connor and Aylward, 2006). This lower binding affinity is fully expressed in human cells and organisms at the level of enzyme induction, which requires approximately 10-fold higher cell or tissue concentrations in humans than in the most sensitive rodent species (reviewed in Connor and Aylward, 2006).
Nohara et al. (2006) observed EC50 values for the expression of CYP1A1 mRNA in human lymphocytes of 1.43 nm and in SD rat lymphocytes of 0.14 nm, indicating that for this end point, humans are about 10-fold less sensitive than SD rats. However, human lymphocytes showed a greater maximal expression of CYP1A1 mRNA at the maximum dose.
Xu et al. (2000) compared the induction of dioxin-induced CYP1A1 activity measured by ethoxyresorufin-O-deethylase (EROD) in primary hepatocyte cultures from humans and SD rats. Rat hepatocytes showed EROD induction at concentrations of 10–12 to 10–10M TCDD, whereas human hepatocytes from five donors were unresponsive at this concentration. The human hepatocytes did not respond at concentrations less than 10–9M TCDD. Maximal induction of EROD was 20-fold that of the solvent control in rat hepatocytes and two- to fivefold in human hepatocytes. Hence, in this study, human hepatocytes display both a lower sensitivity and a lower responsiveness to TCDD than do rat hepatocytes.
Schrenk et al. (1995) measured the potency of TCDD for EROD in hepatocytes from six human donors. The mean EC50 value was 10–10M, about 10-fold higher than that measured in rat hepatocytes (Schrenk et al., 1991).
Silkworth et al. (2005) demonstrated that fresh human hepatocytes from five human donors were about an order of magnitude less sensitive to TCDD than were freshly isolated rat hepatocytes measured by EROD induction. The human hepatocytes were between 100- and 1000-fold less sensitive than rat hepatocytes when measured by CYP1A1 mRNA induction.
Connor and Aylward (2006) provided the most comprehensive review of the measured differences between animals and humans in their sensitivity to dioxin-like chemicals. The data summarized in this review suggest that humans are about 10-fold less sensitive to dioxin-like chemicals than are rodents because of differences in their Ah receptor amino acid composition and structure.
The weight of evidence suggests that humans are about 10-fold less sensitive than SD rats to the effects of DLCs. Hence, a value of 0.1 was used for the toxicodynamic interspecies extrapolation factor, Ufa-TD.
- 2. UFh: This extrapolation factor typically accounts for uncertainties regarding toxicokinetic and toxicodynamic sensitivity within humans. For toxicokinetic variation, a factor of 3 is applied. This factor is the standard UF component assigned to toxicokinetic variability within humans and is consistent with data demonstrating variations ranging from two- to threefold between the median and 95th percentile measured serum lipid concentrations within a given age range for overall dioxin toxic equivalency (Ferriby et al., 2006) and 4-PeCDF in particular (see, e.g., the University of Michigan Dioxin Exposure Study; Garabrant et al., 2007; UMDES, 2006). Hence, we chose a value of 2.5 for UFh-TK.
Toxicodynamic variation in humans has been measured by TCDD induction of EROD in peripheral blood lymphocytes (van Duursen et al., 2005). There was about a sevenfold variation in EC50 for EROD induction. Chang et al. (2007) used immunohistochemical methods to demonstrate variation in AhR, CYP1B1, and CYP1A1 expression in adenocarcinomas from lung cancer patients. In five human adenocarcinoma and squamous cell carcinoma cell lines, there was a fourfold variation in efficacy of the expression of both CYP1A1 and CYP1B1 proteins in response to 1nM TCDD. Chen et al. (2006) examined 225 Taiwanese and determined that serum concentrations of polychlorinated dibenzodioxins and furans and liver function measured by serum enzymes were related to polymorphisms in the CYP1A1/Msp 1 gene. There was about a half-fold difference in serum enzyme activities between the various polymorphic groups, likely reflecting potential differences in efficacy. Roos et al. (2006) studied sedimented urothelial cells from four healthy males aged 30–53 years and observed that CYP1B1 transcription varied about eightfold and CYP1A1 transcription varied about 10-fold in untreated cultured cells. Silkworth et al. (2005) demonstrated about a sevenfold difference in EC50 for EROD induction in human hepatocytes from five donors with a less than twofold difference in efficacy.
Co-exposure to other environmental chemicals may also be a source of toxicodynamic variation in humans. In HepG2 cells, co-exposure to TCDD and mercury or co-exposure to TCDD and lead decreased the efficacy of EROD induction over exposure to a single concentration of TCDD alone; however, co-exposure to TCDD and copper increased the efficacy of EROD induction (Korashy and El-Kadi, 2008). All variations were within twofold. Chao et al. (2006) found that in human hepatoma Huh7 cells, exposure to arsenic reduced TCDD-induced dioxin-responsive element-mediated chemical activated luciferase expression activity by about fivefold and CYP1A1 activation measured by EROD by about 10-fold. No dose-response modeling was performed.
The fold change values for both sensitivity and efficacy related to human variation are 7, 4, 0.5, 8, and 7. The fold change values for efficacy related to co-exposure are 2, 5, and 10. The geometric mean of these eight values is 4. This value was used as UFh-TD.
Combining the values of UFh-TK and UFh-TD, UFh becomes 10. Hence, in this case, the data-derived extrapolation factor and the default UF have the same value.
As an evaluation of the protectiveness of the intraspecies extrapolation factor, a number of in vivo experiments of caffeine metabolism in humans were examined. Caffeine metabolism is a marker for activity of CYP1A2, the dominant CYP1A isozyme in human liver (Butler et al., 1989). Variation in caffeine metabolism rates may be assessed by quantification of exhaled radiolabeled caffeine metabolites, by measurement of the urinary caffeine metabolite ratio (CMR), or by measurement of the caffeine/paraxanthine ratio in plasma (Koch et al., 1999). Because these experiments are performed in vivo, they provide a measure of the combined intraspecies extrapolation factor. ten Tusscher et al. (2008) observed a fivefold variation in the caffeine/paraxanthine ratio in plasma in 37 Dutch children with documented prenatal and lactational exposure to DLCs. Aklillu et al. (2003) measured the urinary CMR in 173 individuals of Ethiopian descent and noted an approximately sevenfold variation in the median values from the various CYP1A2 genotypes found by sequencing DNA from whole blood samples. Smokers and nonsmokers were included. Ghotbi et al. (2007) measured the caffeine/paraxanthine ratio in plasma in 194 Swedes and 150 Koreans. The maximum variation in the ratio was about eightfold. There appeared to be about an eightfold variation in exhaled radiolabeled caffeine metabolites in a group of Yucheng poisoning victims. In the normal controls, the variation was less than fourfold (Lambert et al., 2006). In coastal fish consumers on the north shore of the St Lawrence River, the variation in CMT measured by breath analysis appeared to be about fivefold (Ayotte et al., 2005). Hence, the selection of a value of 10 for the interspecies extrapolation factor is protective based on documented in vivo variation in CYP1A activity in humans.
- 3. UFl: This factor is typically used to account for extrapolation from a POD derived from a lowest observed adverse effect level (LOAEL) to a value presumed to be consistent with a no observed adverse effect level (NOAEL). The range of BMRs modeled here, 1% to 10% tumor incidence, could be regarded as spanning between a NOAEL and a LOAEL. No increase in tumor incidence was observed experimentally even at tissue concentrations above the EC01; however, at the EC10, an increase in tumor incidence was apparent. Selection of the EC01 as a NOAEL would suggest a conclusion that no increase in cancer incidence was likely at exposures below this concentration.
One or more noncancer toxic events may serve as a sentinel or precursor events for tumor formation. The choice of an event as a sentinel is not an easy one because neither of the following is known: 1) the exact sequence of events occurring during the progression of tumor formation in the rats' livers in response to dioxin; and 2) the events that are necessary and/or sufficient for tumor formation in the rats' livers in response to dioxin. The NTP bioassay report presents the end point of toxic hepatopathy and defines it as including all nonneoplastic changes. The fraction of animals displaying toxic hepatopathy at all the doses is provided. Between the presumptive EC01 and EC10, the incidence of toxic hepatopathy increases from around 6% to more than 60% (Fig. 9).
|
Additional information from the bioassay regarding nonneoplastic hepatic toxicity indicates that hepatocyte hypertrophy, diffuse fatty change, and increased pigmentation all show sharp increases at tissue concentrations below the EC01, whereas oval cell hyperplasia and necrosis show increases at similar or higher doses than the tumor response (NTP, 2006a
).
CYP induction, as measured by EROD activity or the occurrence of 
-glutamyl transpeptidase-positive altered hepatic foci (GT+-AHF), was also considered as continuous end point that might reflect tumor formation (NTP, 2006a; Waern et al., 1991) (Fig. 9). At the lowest bioassay dose, the liver concentrations were approximately three- to fourfold lower than the LEC01, and little or no hepatic toxicity as measured by toxic hepatopathy was apparent. The largest increases in EROD induction and the occurrence of GT+-AHF occurred between the EC01 and EC10.
POD levels for toxic hepatopathy, EROD induction, and GT+-AHF occurrence were also developed to determine if the POD levels for these noncancer end points would be similar to the EC01 and EC10 for the combined tumor response. Similarity would support the choice of the EC01 and EC10 of the combined liver tumor response as representative of the NOAEL and LOAEL, respectively (USEPA, 2005).
The identified UFl is applied to the modeled tissue concentrations in rats and its value is 10 when applied to the EC10 and 1 when applied to the EC01 (see Fig. 1).
- 4. UFs: This factor, intended to account for less-than-chronic study durations, was judged not appropriate because the NTP bioassays were lifetime studies (104 weeks); therefore, a UFs of 1 was used.
- 5. UFd: This factor is applied in risk assessments to acknowledge uncertainty based on limited available data sets. However, in the case of dioxin-like compounds, an extensive database is available, so a UF of 1 was applied in this case.
- 5. UFd: This factor is applied in risk assessments to acknowledge uncertainty based on limited available data sets. However, in the case of dioxin-like compounds, an extensive database is available, so a UF of 1 was applied in this case.
The identified extrapolation factors were applied to the benchmark concentrations in rats as depicted in Figure 1 to estimate a human RfD based on the cancer end point.
Figure 2 illustrates the approach used to estimate cancer potency factors based on a linear extrapolation from the modeled POD. The lower confidence limit HEDs (LHEDs) were back-extrapolated from the LECx tissue concentrations in rats using the Carrier toxicokinetic model. Cancer potency factors were calculated by linear extrapolation from the POD values expressed as LHEDs by calculating the ratio between the BMR rate (i.e., 1% or 10% noted as "x") and the corresponding LHED:
 |
|
Traditional RfD approach for comparison.
RfDs and cancer slope factors (CSFs) were also developed based on a more traditional approach for comparison. BMD values in rats were estimated using the average tissue concentrations developed with the trapezoidal method according to Gray et al. (2006)
. First, the rat external doses were multiplied by 5/7 to convert them to seven-days-a-week external dose values. Linear fits of the average tissue concentrations to these seven-day-a-week external dose values were used to obtain external doses in rats corresponding to the various BMD values or their lower confidence limit on the benchmark dose.
For RfD derivation, UFl and UFa were applied to these external dose values corresponding to the BMD at the PODs. Allometric scaling using the 
power of the body weight ratio was then used to convert these doses to human equivalents. The choice of a scaling factor of is appropriate for orally administered chemicals in which the toxicity is attributable to the parent chemical, as is the case for 4-PeCDF. However, the use of allometric scaling is, in general, not as credible a method for interspecies extrapolation as is the use of toxicokinetic modeling, especially for persistent chemicals (Kirman et al., 2003; USEPA, 2006). UFh was applied to these human equivalent BMDx values to obtain RfD values.
For CSFs, the external doses in rats corresponding to the BMDL values were allometric scaled to obtain human equivalents. The human equivalents were divided into the risk level at each POD to obtain CSF values.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Evaluation/Validation of Rat and Human Toxicokinetic Models
Evaluation of the Carrier Model for the Rat
The Carrier model provided a good fit for rat tissue concentration data available in the 2006 NTP study for 4-PeCDF (Figs. 3 and 4). The modeled average tissue concentrations were very similar to the measured average tissue concentrations calculated from the NTP data using the method of Gray et al. (2006)
and the trapezoidal rule. Table 2 shows both the modeled tissue concentrations and those estimated with the trapezoidal rule. Overall, the model provides an excellent fit to the NTP bioassay data. However, there was a tendency to underestimate the liver concentration at week 31 across all dose groups and to underestimate adipose concentration at the two lowest doses toward the end of the exposure period. The highest measured concentrations in the rats' liver occurred at 31 weeks, whereas the highest measured concentration in the rats' adipose tissue occurred at 104 weeks. In contrast, the modeled concentrations in both liver and adipose tissue increased monotonically. What these observations suggest is that during the first 30 weeks of the bioassay, 4-PeCDF accumulated in the rats' liver to a greater extent than the model predicted and at later times redistributed to adipose tissue.
Evaluation of the Carrier Model for Humans
The Carrier model has been applied to human data since its creation (Aylward et al., 2004, 2005; Carrier et al., 1995b). In previous studies, the Carrier model was demonstrated to simulate very high human exposures, such as those occurring in Yusho or Yucheng victims, the two Austrian poisoning victims, Seveso victims, or the chemical plant workers in the National Institute for Occupational Safety and Health cohort (Aylward et al. 2005; Carrier et al. 1995a,b).
Model output based on estimates of dietary intakes of 4-PeCDF was compared with liver and adipose tissue data from Japanese experiencing background exposure (Iida et al., 1999; Maruyama et al., 2002). The model reproduced the data quite well (Fig. 5).
|
The model was also able to reproduce the decline of lipid-normalized blood concentrations following very high exposures. Figure 6 shows a comparison of the output of each model with blood concentrations of 4-PeCDF measured in five Yusho patients and three Yucheng patients (Masuda, 2001
; Ryan et al., 1993).
|
Tissue Concentrations for Tumor End Points in Rats Obtained from BMD Modeling
The original bioassay administered doses, liver tumor responses (unadjusted and adjusted for survival), and estimated and measured liver and adipose tissue concentrations are presented in Table 2. In BMD modeling, the probit model consistently produced the best fit to the data. Other models, such as the logistic and multistage models, provided results for both dose metrics that were within 12% at the EC10 and within 5% at the EC01. Fitting statistics for all models are provided in the Supplementary Information. Figure 7 displays the dose-response data and BMD modeling results graphically, and Table 3 presents the estimated benchmark tissue concentrations and confidence bounds at the 1% and 10% response levels from both the modeled and the measured lifetime average tissue concentrations.
|
|
The modeled EC01 tissue concentrations and their lower bounds were well within the experimental dose range. The 10% response benchmark concentrations and their lower bounds were approximately four to five times greater than those for the 1% response level, reflecting the nonlinearity in the dose-response data at these concentrations. All benchmark concentration estimates were robust, with lower bound estimates generally within 20% of the central estimates and the upper bound estimates all within a factor of 3 of the lower bound estimates. Based on these results, the EC10 and EC01 estimates were used for development of RfDs and the LEC10 and LEC01 estimates were carried forward for estimation of cancer potency factors.
Using the measured liver concentrations and the nonzero value for the vehicle control group raised the BMDs by 27% and 26% at the 1% and 10% PODs, respectively. Using the measured adipose concentrations and the LOQ for the vehicle control group raised the BMDs by 19% and 0.7% at the 1% and 10% PODs, respectively (Table 5).
|
Application of Extrapolation Factors to Obtain HEDs and Toxicity Criteria
Table 4 presents a summary of the extrapolation factors used in the derivation of RfDs based on the BMD modeling and the application of the toxicokinetic model. Table 5 details the derivation of the RfDs. External HEDs were obtained using the human Carrier model after applying the UFl and UFa extrapolation factor components to internal dose metrics (EC01 values for LALC and LAAC) as illustrated in Figure 1; the UFh component was applied to the estimated external human doses to derive the RfDs. The RfDs in nanogram per kilogram per day presented for each dose metric and POD value model reflect the range of UFa values as discussed above and presented in Table 5.
|
When compared with the traditional approach, both the choice of data-derived extrapolation factors and the use of the toxicokinetic model make a big difference in the values of the RfDs. These differences are not simple.
When LALC is used as the dose metric, application of data-derived UFs increases the value of the RfD between one and two orders of magnitude (0.07 vs. 1 ng/kg/day at the 1% POD; 0.04 vs. 2 ng/kg/day at the 10% POD). The use of the toxicokinetic model for species extrapolation increases the RfD values by three orders of magnitude (2000 vs. 1 ng/kg/day at the 1% POD; 1000 vs. 2 ng/kg/day at the 10% POD) (Table 5).
When LAAC is used as the dose metric, application of data-derived UFs increases the value of the RfD one to two orders of magnitude (0.06 vs. 2 ng/kg/day at the 1% POD; 0.04 vs. 2 ng/kg/day at the 10% POD). The use of the toxicokinetic model for species extrapolation increases the RfD values by one to two orders of magnitude (30 vs. 2 ng/kg/day at the 1% POD; 300 vs. 2 ng/kg/day at the 10% POD) (Table 5).
Table 6 presents the derivation of the CSFs based on the modeled LEC10 and LEC01 liver and adipose lifetime average tissue concentrations according to the approach outlined in Figure 2. Based on these CSFs, risk-specific doses (RSDs) associated with a cancer risk range of 10–6 to 10–4 can be estimated and used as a basis for comparison to the RfDs (Fig. 8).
|
|
In contrast to the large differences in the RfD values, the CSFs derived from measured tissue concentrations and allometric scaling are quite similar in value to those derived using the toxicokinetic model for interspecies extrapolation.
Figure 8 illustrates the range of RSDs and RfDs based upon different choices of tissue and toxicokinetic model. The range of RfDs derived based on the POD/UF approach is between three and four orders of magnitude above the range of RSDs associated with 10–6 to 10–4 cancer risk based on linear extrapolation from the POD, and the full range of the potentially "tolerable" doses derived through this analysis spans almost nine orders of magnitude. In contrast, the RSD at a risk of 10–6 calculated from the 1989 USEPA dioxin assessment is 6 x 10–6 ng/kg/day and that from the 2003 dioxin reassessment is 1 x 10–6 ng/kg/day, both below the lower end of the range of RSDs derived here (USEPA, 1989, 2003).
Results from the Use of Traditional Approaches for Developing Toxicity Criteria
In addition to the nonlinear cancer toxicity criteria developed with pharmacokinetic modeling, BMD modeling, and data-derived UFs, Table 5 also shows RfDs developed using measured concentrations in rats in lieu of modeled concentrations, allometric scaling for species extrapolation in lieu of pharmacokinetic modeling, and application of default UFs. These criteria derived using more traditional methods are presented for comparison purposes only, and the preferred criteria are those in the upper portion of Table 5 and derived in a fashion consistent with Environmental Protection Agency's Cancer Guidelines (USEPA, 2005).
The use of the allometric scaling with data-derived extrapolation factors reduces the derived RfD values by between one and three orders of magnitude. The use of default values for extrapolation factors reduces the derived RfD values by more than an order of magnitude (Table 5).
The use of allometric scaling versus the toxicokinetic modeling produced less than an order of magnitude change in the values of the CSFs (Table 6).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
The cancer end point was chosen here for consistency with previous risk assessments of dioxin-like chemicals. It is important to note that cancer may not be an appropriate end point for humans (Cole et al., 2003
). That said, the quantitative risk assessment results presented here highlight several key elements that are considered in risk assessment decisions, namely the modeling approach and the internal dose metric.
Linear versus Nonlinear Approach
The decision with the greatest quantitative impact is the choice of a linear versus nonlinear approach to cancer risk assessment. This decision hinges upon the carcinogenic MOA for 4-PeCDF.
The likely carcinogenic MOA for 4-PeCDF and other AhR agonists in the liver is tumor promotion of spontaneously initiated hepatocytes that occur with threshold-dependent characteristics (Pitot et al., 1987; Viluksela et al., 2000). TCDD and 4-PeCDF are not mutagenic or genotoxic; therefore, their MOA is not one of a direct acting mutagen or initiator (NTP, 2006a,b; Randerath et al., 1988, 1990, 1993; Shu et al., 1987; Slikker et al., 2004a,b; Turteltaub et al., 1990; Watson et al., 1995). Rather, a likely MOA for DLCs includes tumor promotion of spontaneously initiated foci and late-stage, high-dose hepatopathy resulting in increased regenerative-repair-induced cell division (Graham et al., 1988; Hailey et al., 2005; Pitot et al., 1987; Teeguarden et al., 1999; Viluksela et al., 2000). Key events underlying the MOA include binding to and activation of the AhR, inhibition of apoptosis in spontaneously initiated foci, increased foci proliferation, and increased cell division (Conolly and Andersen, 1997; Hailey et al., 2005; Kociba et al., 1978; NTP, 2006a,b; Schrenk et al., 2004; Schwarz and Appel, 2005; Stinchcombe et al., 1995; Teeguarden et al., 1999; Waern et al., 1991). Regenerative proliferation in response to hepatopathy occurs late in the process and only after tissue accumulation of 4-PeCDF has occurred (NTP, 2006a). Significant tumor incidence does not occur until after hepatopathy develops and, thus, requires high levels of 4-PeCDF accumulation. These key events and MOAs exhibit dose-response and temporality relationships that are consistent with the overall coherence of a large body of evidence published on AhR ligands such as 4-PeCDF. Furthermore, the MOAs and key events are biologically plausible with what is known about the multistage model for liver cancer, including the MOAs and key events for other nuclear receptor ligands such as phenobarbital and peroxisome proliferator activated receptor alpha ligands (Holsapple et al., 2006; Roberts et al., 1997).
A human relevance framework for applying the knowledge of the MOA to risk assessment has been developed by a number of regulatory and scientific organizations (Cohen et al., 2004; Holsapple et al., 2006; Meek et al., 2003; Seed et al., 2005; Sonich-Mullin et al., 2001; USEPA, 2005). The MOA and key events for AhR ligands such as 4-PeCDF, when examined with the criteria of the human relevance framework, provide strong scientific support for applying a threshold basis toxicity criterion such as an RfD to the cancer risk assessment of 4-PeCDF. The use of a threshold for 4-PeCDF is consistent with USEPA's Cancer Guidelines and the recommendations made by the NAS in their review of the USEPA's dioxin risk assessment (NAS, 2006; USEPA, 2005).
Sentinel Events as Precursors to Tumor Formation
The possibility that an event reflecting noncancer toxicity could be used as the basis of an RfD protective of cancer was also explored. A precursor toxic effect could also be associated with a key event underlying the carcinogenic MOA. This choice provides flexibility, especially for nonlinear dose-response modeling (USEPA, 2005). Hence, three end points of AhR activation were considered: (1) toxic hepatopathy, (2) induction of CYP as measured by EROD at 53 weeks of the NTP bioassay, and (3) the occurrence of GT+-AHF (NTP, 2006a; Waern et al., 1991).
The most difficult choice in modeling a noncancer end point in this way is the choice of POD. The choice of POD should reflect a measurable difference in the dose-response for the selected noncancer event and should also provide adequate protection for the cancer end point.
Toxic hepatopathy was reported as a dichotomous response (NTP, 2006a). Regarding the choice of POD for the toxic hepatopathy end point, the highest observed frequency of toxic hepatopathy (83%) was about fivefold that of the combined tumor response (16.2%) (Fig. 9). Hence, one could assume that the EC50 and the EC05 for toxic hepatopathy would correspond to the EC10 and EC01 for the combined tumor response. Using this end point, the EC50 and EC05 for LALC were 209,100 and 36,500 ng/kg, respectively. The RfDs ensuing from the use of toxic hepatopathy as an end point and LALC as the dose metric would be 30–50% lower than those derived using combined liver tumors as the end point and LALC as the dose metric. Given the other uncertainties inherent in this assessment, this difference is not large. Hence, the dose-response for toxic hepatopathy fulfills the dose-response criteria for the human relevance framework and supports the use of the POD values derived for combined liver tumors (Cohen et al., 2004; Meek et al., 2003; USEPA, 2005).
EROD activity reflects induction of CYP induction, and the possibility of using this as an indicator of AhR receptor activation was also examined. EROD induction measures AhR activation but is not necessarily linked to toxicity. Figure 9 shows the relationship between the dose-response for EROD induction at 53 weeks and the combined liver tumor response. The Hill model has been used previously to evaluate the relative potency factor for 4-PeCDF and was also used to obtain BMD values (Toyoshiba et al., 2004). Details of this model fit are provided in the Supplementary Information. The largest fold increase in percent EROD induction response occurs between external doses of 20 and 92 ng/kg/day. Mean EROD induction was about 1000 pmol/min/mg protein and about 3000 pmol/min/mg external doses of 20 and 92 ng/kg/day, respectively. Using these values as BMR levels for continuous modeling of EROD induction gives BMD values of 44,200 and 278,300 ng/kg LALC, respectively. Respectively, these POD values are 60% and 90% of the EC01 and EC10 values based on combined tumor response (Table 3). Hence, dose-response evaluation for EROD induction also supports the use of POD values based on combined liver tumors.
The occurrence of GT+-AHF could also serve as a tumor precursor event in a more direct fashion than the other two noncancer end points. Waern et al. (1991) measured the number of GT+-AHF per liver in female SD rats given 160, 640, or 2600 ng/kg/week 4-PeCDF for 20 weeks in weekly subcutaneous injections as a tumor promoter after initiation with nitrosodiethylamine. 4-PeCDF increased both the number and the volume fraction of GT+-AHF in a dose-dependent fashion. Estimates of LALC were obtained with the Carrier model assuming that all the weekly injection of 4-PeCDF was absorbed and averaging the modeled liver concentration over 20 weeks. Figure 9 shows the dose-response relationship of the number of GT+-AHF with LALC compared with the combined tumor response. The Power model gave the best fit to the data, and statistics are provided in the Supplementary Information. A value of 3000 GT+-AHF corresponded to the EC01 and a value of 4000 GT+-AHF corresponded to the EC10. Both these AHF values were within the confidence limits of the response at the 640 ng/kg/week dose. The BMD values associated with the AHF values of 3000 and 4000 were 74,800 and 301,100 ng/kg LALC, respectively. These values are almost identical with the EC01 and EC10 values for the combined tumor response. The occurrence and size of AHF are believed to be related to promotional activity and, as such, are very likely the noncarcinogenic end points that are most closely related to tumor formation (Waern et al., 1991).
Consideration of all three noncancer end points shows that the dose-response of key events within the MOA is similar to the dose-response of combined liver tumors and links these key events to the carcinogenic outcome. The differences between BMDs obtained with these three noncancer end points are vanishingly tiny when compared to the differences ensuing from the use of a nonlinear versus linear approach, the choice of LAAC as the dose metric, or the use of allometric scaling and default values for extrapolation factors.
Regarding the issue of nonlinear versus linear dose-response evaluation, reducing the LALC-based RfD values in Table 5 by 50% still puts them orders of magnitude above the RSDs shown in Table 6.
The Choice of Dose Metric
The second decision that substantially affects the resulting toxicity criteria is the choice of internal dose metric, and particularly, the decision of whether to rely upon liver or adipose tissue concentrations. In rodents, 4-PeCDF is sequestered in the liver by binding to induced CYP1A2 protein (Diliberto et al. 1999), but it is not known whether CYP1A2-bound 4-PeCDF is biologically active. Clearly, induced hepatic binding, sequestration, and elimination play a role in the potential toxicity of 4-PeCDF and other dioxin-like chemicals.
Furthermore, it is uncertain to what degree such induced binding, sequestration, and elimination may play a role in human liver or other tissues at various levels of exposure when compared to the role of binding, sequestration, and elimination in rodents. In short, the toxicokinetic details of tissue distribution and the toxicodynamic details of reversible binding, tissue sequestration, and elimination may prove to be significant factors in the species differences in the toxicity of both 4-PeCDF and TCDD. This uncertainty is highlighted by the apparent hepatic accumulation of 4-PeCDF in the rats' liver at 31 weeks followed by apparent redistribution to adipose tissue to 104 weeks. The uncertainties and complexities related to this issue complicate the choice of best internal dose metric for use in risk assessment.
Given the lack of any strong tumor responses in nonhepatic tissues, the adipose tissue concentration is not an appropriate dose metric to use in conjunction with the liver tissue tumor response: if the adipose tissue concentration represents a marker of "free" or biologically active compound that is suitable for assessing the dose-response for hepatic tissues, then tumor responses might have been expected in other nonhepatic tissues as well. With the exception of a borderline significant response at the highest dose level in gingival tissue, the notable lack of nonhepatic tumor responses indicates that adipose tissue concentrations do not provide a representative dose metric for hepatic cancer dose-response assessment. For this reason, LALC is the most appropriate dose metric for dose-response assessment of rodent liver tumors due to exposure to 4-PeCDF, in spite of any uncertainties regarding the impact of binding to CYP1A2 protein.
The uncertainty associated with the use of the LAAC as a dose metric is highlighted by the order of magnitude difference between the RfD values at the 1% and 10% PODs (30 vs. 300 ng/kg/day), whereas the corresponding RfDs based on LALC differ by only a factor of 2 (2000 vs. 1000 ng/kg/day) (Table 5). What this variation and lack of consistency suggest is that adipose tissue concentration is not a suitable dose metric for liver tumors.
Toxicokinetic Modeling versus Allometry and Measured Tissue Concentrations
Based on the two choices discussed above, use of a nonlinear (RfD) approach and use of LALC as the internal dose metric of interest, a range of RfD estimates were calculated. The RfD range is from 0.04 to 2000 ng/kg/day depending on whether one used default extrapolation factors or data-derived extrapolation factors or whether one chose to use a toxicokinetic model and a toxicodynamic extrapolation factor or allometric scaling for animal-to-human extrapolation.
Initially, the Maruyama model was considered for this work. However, we wished to use the same toxicokinetic model for both rats and humans. The Carrier model incorporates a concentration-dependent increase in distribution to hepatic tissue at increasing doses. The Maruyama model does not include any dose-dependent changes in tissue distribution patterns. It is notable that Maruyama and Aoki (2006) used a model by Andersen et al. (1993) to estimate tissue concentration in rats. The Andersen model includes dose-dependent hepatic binding and accumulation. Presumably, the Maruyama model has difficulties in rats because of its lack of dose-dependent hepatic accumulation. Both the USEPA cancer guidelines and the scientific literature suggest that the use of the same toxicokinetic model in both animals and humans is best practice in risk assessment (Clewell et al., 2002; USEPA, 2005). Some preliminary calculations with the human Maruyama model suggested that LALC-based RfD values could be about 10-fold higher than those shown in Table 5.
Regarding the use of modeled versus measured concentrations, in all cases, BMDs based on measured tissue concentrations were only slightly higher than those based on modeled concentrations, and using the modeled tissue concentrations is a protective choice for risk assessment (Table 3). The choice of using the modeled tissue concentrations or measured tissue concentrations or a measured concentration for the vehicle control group contributed little to the overall uncertainty in the toxicity criteria.
Comparison with Regulatory Criteria
These RfD estimates based on rat liver tumors are four to five orders of magnitude higher than estimated tolerable daily intakes of TCDD-toxic equivalents (TEQ) based on various noncancer end points. For example, the WHO Joint FAO/WHO Expert Committee on Food Additives (JECFA) established a tolerable monthly intake rate for TEQ equivalent to 2.3 pg TEQ/kg/day (WHO, 1998, 2000). For 4-PeCDF at the current WHO TEF value, this would correspond to approximately 7 pg/kg/day, about five orders of magnitude lower than the LALC-based RfD at either the 1% or the 10% POD. The value of 7 pg/kg/day is about twofold lower than the RSD at a risk of 10–4 based on LALC-derived CSF at the 1% POD and about 14-fold lower than the RSD at 10–4 risk calculated from the LALC-derived CSF at the 10% POD.
The WHO JECFA value of 7 pg/kg/day is based on decreased sperm counts, immune suppression, and increased frequency of genital malformations in rat offspring, endometriosis in monkeys, and neurodevelopmental effects in monkey offspring (WHO, 1998, 2000). The conclusion of WHO JECFA that cancer may not be the most sensitive end point observed in laboratory animals that also has relevance for humans is supported by the fact that the cancer-based RfDs derived here are orders of magnitude higher than 7 pg/kg/day.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
A comprehensive cancer bioassay was recently published for 4-PeCDF, obviating the need to rely on a TEF for risk assessment of this chemical (NTP, 2006a
). This situation is in contrast to that of most of the other seventeen 2,3,7,8-substituted chlorinated dibenzodioxins and furans for which no cancer bioassay data exist, thus resulting in the reliance on TEFs when conducting a risk assessment. As an "interim" regulatory policy, TEFs possess limitations and uncertainty in several areas (NAS, 2006). Some of the factors contributing to the uncertainty underlying the TEF values and approach in application to cancer risk assessment include (1) the wide range of in vivo and in vitro end points (some of which may not be relevant to cancer responses) and resulting estimates of relative potencies supporting each TEF, (2) the lack of a rigorous method for selecting a TEF value from the available range of potency estimates for a given congener, and (3) a lack of consideration of potential interspecies differences in the pattern of toxicokinetics and toxicodynamic responses among the various congeners.
For these and other reasons, when risk assessments are dominated by exposures other than TCDD and when relevant congener-specific toxicity data are available, those data should be given precedence over, or at a minimum, given equal weight to risk assessments based on the traditional TEF approach and TCDD-specific toxicity data. The risk assessment presented here incorporates elements recommended by both the NAS (2006) and the USEPA's Science Advisory Board (USEPA SAB, 2001) in their recent reviews of the USEPA dioxin reassessment drafts (USEPA, 1989, 2003): (1) use of a nonlinear approach, (2) reliance on internal dose metrics, (3) accounting for toxicokinetic differences in absorption, distribution, and elimination between humans and rats, and (4) a quantitative assessment of the uncertainties associated with various choices in the risk assessment process.
This quantitative uncertainty analysis demonstrated that choice of approach (linear vs. nonlinear) had the largest impact on estimated tolerable doses, with estimated tolerable doses approximately 1000-fold or more lower based on the linear approach depending on the RSD chosen. The choice of whether to use a toxicokinetic model or allometric scaling accounts for a 10- to 1000-fold difference in the RfD values depending on the dose metric or POD value. The choice of whether to use data-derived extrapolation factor values or default values in the RfD derivation also accounted for a change of about 100-fold in the RfD values.
Based on MOA data, we recommend a nonlinear approach relying upon liver tissue concentration as the internal dose metric of interest as the basis of a scientifically justifiable cancer RfD for 4-PeCDF.
|
SUPPLEMENTARY DATA |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Supplementary data are available online at http://toxsci.oxfordjournals.org/.
|
FUNDING |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Preparation of this manuscript was supported by funding from the Dow Chemical Company.
|
ACKNOWLEDGMENTS |
|---|
The manuscript is dedicated to the memory of Thomas F. Long, a valued colleague and good friend who contributed to this effort but unfortunately passed away before its completion.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Aklillu E, Carrillo JA, Makonnen E, Hellman K, Pitarque M, Bertilsson L, Ingelman-Sundberg M. Genetic polymorphism of CYP1A2 in Ethiopians affecting induction and expression: Characterization of novel haplotypes with single-nucleotide polymorphisms in intron 1. Mol. Pharmacol. (2003) 64(3):659–669.
Andersen ME, Mills JJ, Gargas ML, Kedderis L, Birnbaum LS, Neubert D, Greenlee WF. Modeling receptor-mediated processes with dioxin: Implications for pharmacokinetics and risk assessment. Risk Anal. (1993) 13(1):25–36.[CrossRef][Web of Science][Medline]
Aylward LL, Brunet RC, Carrier G, Hays SM, Cushing CA, Needham LL, Patterson DG Jr, Gerthoux PM, Brambilla P, Mocarelli P. Concentration-dependent TCDD elimination kinetics in humans: Toxicokinetic modeling for moderately to highly exposed adults from Seveso, Italy, and Vienna, Austria, and impact on dose estimates for the NIOSH cohort. J. Expo. Anal. Environ. Epidemiol. (2004) 15:51–65.[CrossRef][Web of Science]
Aylward LL, Brunet RC, Starr TB, Carrier G, Delzell E, Cheng H, Beall C. Exposure reconstruction for the TCDD-exposed NIOSH cohort using a concentration- and age-dependent model of elimination. Risk Anal. (2005) 25:945–956.[CrossRef][Web of Science][Medline]
Ayotte P, Dewailly E, Lambert GH, Perkins SL, Poon R, Feeley M, Larochelle C, Pereg D. Biomarker measurements in a coastal fish-eating population environmentally exposed to organochlorines. Environ. Health Perspect. (2005) 113(10):1318–1324.[Web of Science][Medline]
Birnbaum LS, DeVito MJ. Use of toxic equivalency factors for risk assessment for dioxins and related compounds. Toxicology (1995) 105:391–401.[CrossRef][Web of Science][Medline]
Brown RP, Delp MD, Lindstedt SL, Rhomberg LR, Beliles RP. Physiological Parameter Values for physiologically-based pharmacokinetic models. Toxicol. Ind. Health (1997) 13:407–484.
Budinsky RA, Paustenbach D, Fontaine D, Landenberger B, Starr TB. Recommended relative potency factors for 2,3,4,7,8-pentachlorodibenzofuran: The impact of different dose metrics. Toxicol. Sci. (2006) 91:275–285.
Butler MA, Iwasaki M, Guengerich FP, Kadlubar FF. Human cytochrome P-450PA (P-450IA2), the phenacetin O-deethylase, is primarily responsible for the hepatic 3-demethylation of caffeine and N-oxidation of carcinogenic arylamines. Proc. Natl. Acad. Sci. U.S.A. (1989) 86(20):7696–7700.
Carrier G, Brunet RC, Brodeur J. Modeling of the toxicokinetics of polychlorinated dibenzo-p-dioxins and dibenzofurans in mammalians, including humans. I. Nonlinear distribution of PCDD/PCDF body burden between liver and adipose tissues. Toxicol. Appl. Pharmacol. (1995a) 131:253–266.[CrossRef][Web of Science][Medline]
Carrier G, Brunet RC, Brodeur J. Modeling of the toxicokinetics of polychlorinated dibenzo-p-dioxins and dibenzofurans in mammalians, including humans. II. Kinetics of absorption and disposition of PCDDs/PCDFs. Toxicol. Appl. Pharmacol. (1995b) 131:267–276.[CrossRef][Web of Science][Medline]
Chang JT, Chang H, Chen PH, Lin SL, Lin P. Requirement of aryl hydrocarbon receptor overexpression for CYP1B1 up-regulation and cell growth in human lung adenocarcinomas. Clin. Cancer Res. (2007) 13(1):38–45.
Chao HR, Tsou TC, Li LA, Tsai FY, Wang YF, Tsai CH, Chang EE, Miao ZF, Wu CH, Lee WJ. Arsenic inhibits induction of cytochrome P450 1A1 by 2,3,7,8-tetrachlorodibenzo-p-dioxin in human hepatoma cells. J. Hazard. Mater. (2006) 137(2):716–722.[CrossRef][Web of Science][Medline]
Chen HL, Su HJ, Wang YJ, Guo YL, Liao PC, Lee CC. Interactive effects between CYP1A1 genotypes and environmental polychlorinated dibenzo-p-dioxins and dibenzofurans exposures on liver function profile. J. Toxicol. Environ. Health A (2006) 69(3–4):269–281.[CrossRef][Web of Science][Medline]
Clewell HJ III, Andersen ME, Barton HA. A consistent approach for the application of pharmacokinetic modeling in cancer and noncancer risk assessment. Environ. Health Perspect. (2002) 110(1):85–93.[Web of Science][Medline]
Cohen SM, Klaunig J, Meek ME, Hill RN, Pastoor T, Lehman-McKeeman L, Bucher J, Longfellow DG, Seed J, Dellarco V, et al. Evaluating the human relevance of chemically induced animal tumors. Toxicol. Sci. (2004) 78:181–186.
Cole P, Trichopoulos D, Pastides H, Starr T, Mandel JS. Dioxin and cancer: A critical review. Regul. Toxicol. Pharmacol. (2003) 38:378–388.[CrossRef][Web of Science][Medline]
Connor KT, Aylward LL. Human response to dioxin: Aryl hydrocarbon receptor (AhR) molecular structure, function, and dose-response data for enzyme induction indicate an impaired human AhR. J. Toxicol. Environ. Health B Crit. Rev. (2006) 9:147–171.[CrossRef][Web of Science][Medline]
Conolly RB, Andersen ME. Hepatic foci in rats after diethylnitrosamine initiation and 2,3,7,8-tetrachlorodibenzo-p-dioxin promotion: Evaluation of a quantitative two-cell model and of CYP 1A1/1A2 as a dosimeter. Toxicol. Appl. Pharmacol. (1997) 146(2):281–293.[CrossRef][Web of Science][Medline]
Cox DR, Oakes D. Analysis of Survival Data. (1984) London, New York: Chapman and Hall.
Diliberto JJ, Burgin DE, Birnbaum LS. Effects of CYP1A2 on disposition of 2,3,7, 8-tetrachlorodibenzo-p-dioxin, 2,3,4,7,8-pentachlorodibenzofuran, and 2,2',4,4',5,5'-hexachlorobiphenyl in CYP1A2 knockout and parental (C57BL/6N and 129/Sv) strains of mice. Toxicol. Appl. Pharmacol. (1999) 159:52–64.[CrossRef][Web of Science][Medline]
Dorne JL, Renwick AG. The refinement of uncertainty/safety factors in risk assessment by the incorporation of data on toxicokinetic variability in humans. Toxicol. Sci. (2005) 86:20–26.
Ferriby LL, Knutsen JS, Harris M, Unice KM, Scott P, Nony P, Haws LC, Paustenbach D. Evaluation of PCDD/F and dioxin-like PCB serum concentration data from the 2001–2002 National Health and Nutrition Examination Survey of the United States population. J. Exp. Sci. Environ. Epidemiol. (2006) 17:358–371.
Garabrant D, Chen Q, Hong B, Franzblau A, Lepkowski J, Adriaens P, Demond A, Hedgeman E, Knutson K, Chang C-W, et al. Logistic regression models for high serum 2,3,7,8-TCDD concentrations in residents of Midland, Michigan, USA. Organohalogen Compd. (2007a) 69:2203–2206.
Garabrant D, Hong B, Chen Q, Franzblau A, Lepkowski J, Adriaens P, Demond A, Hedgeman E, Knutson K, Zwica L, et al. Factors that predict serum dioxin concentrations in Michigan, USA. Organohalogen Compd. (2007b) 69:206–209.
Ghotbi R, Christensen M, Roh HK, Ingelman-Sundberg M, Aklillu E, Bertilsson L. Comparisons of CYP1A2 genetic polymorphisms, enzyme activity and the genotype-phenotype relationship in Swedes and Koreans. Eur. J. Clin. Pharmacol. (2007) 63(6):537–546.[CrossRef][Web of Science][Medline]
Graham MJ, Lucier GW, Linko P, Maronpot RR, Goldstein JA. Increases in cytochrome P-450 mediated 17 beta-estradiol 2-hydroxylase activity in rat liver microsomes after both acute administration and subchronic administration of 2,3,7,8-tetrachlorodibenzo-p-dioxin in a two-stage hepatocarcinogenesis model. Carcinogenesis (1988) 9(11):1935–1941.
Gray MN, Aylward LL, Keenan RE. Relative cancer potencies of selected dioxin-like compounds on a body-burden basis: Comparison to current toxic equivalency factors (TEFs). J. Toxicol. Environ. Health A (2006) 69:907–917.[CrossRef][Web of Science][Medline]
Hailey JR, Walker NJ, Sells DM, Brix AE, Jokinen MP, Nyska A. Classification of proliferative hepatocellular lesions in Harlan Sprague-Dawley rats chronically exposed to dioxin-like compounds. Toxicol. Pathol. (2005) 33:165–174.
Haws LC, Su SH, Harris M, DeVito MJ, Walker NJ, Farland WH, Finley B, Birnbaum LS. Development of a refined database of mammalian relative potency estimates for dioxin-like compounds. Toxicol. Sci. (2006) 89:4–30.
Holsapple MP, Pitot HC, Cohen SM, Boobis AR, Klaunig JE, Pastoor T, Dellarco VL, Dragan YP. Mode of action in relevance of rodent liver tumors to human cancer risk. Toxicol. Sci. (2006) 89:51–56.
Iida T, Hirakawa H, Matsueda T, Nagayama J, Nagata T. Polychlorinated dibenzo-p-dioxins and related compounds: Correlations of levels in human tissues and in blood. Chemosphere (1999) 38:2767–2774.[Medline]
Kirman CR, Sweeney LM, Meek ME, Gargas ML. Assessing the dose-dependency of allometric scaling performance using physiologically based pharmacokinetic modeling. Regul. Toxicol. Pharmacol. (2003) 38:345–367.[CrossRef][Web of Science][Medline]
Kitamura K, Nagahashi M, Sunaga M, Watanabe S, Nagao M. Balance of intake and excretion of 20 congeners of polychlorinated dibenzo-p-dioxin, polychlorinated dibenzofuran and coplanar polychlorinated biphenyl in healthy Japanese men. J. Health Sci. (2001) 47:145–154.[CrossRef]
Koch JP, ten Tusscher GW, Koppe JG, Guchelaar HJ. Validation of a high-performance liquid chromatography assay for quantification of caffeine and paraxanthine in human serum in the context of CYP1A2 phenotyping. Biomed. Chromatogr. (1999) 13(4):309–314.[CrossRef][Web of Science][Medline]
Kociba RJ, Keyes DG, Beyer JE, Carreon RM, Wade CE, Dittenber DA, Kalnins RP, Frauson LE, Park CN, Barnard SD, et al. Results of a two-year chronic toxicity and oncogenicity study of 2,3,7,8-tetrachlorodibenzo-p-dioxin in rats. Toxicol. Appl. Pharmacol. (1978) 46(2):279–303.[CrossRef][Web of Science][Medline]
Korashy HM, El-Kadi AO. Modulation of TCDD-mediated induction of cytochrome P450 1A1 by mercury, lead, and copper in human HepG2 cell line. Toxicol. In Vitro (2008) 22(1):154–158.[CrossRef][Web of Science][Medline]
Lambert GH, Needham LL, Turner W, Lai TJ, Patterson DG Jr, Guo YL. Induced CYP1A2 activity as a phenotypic biomarker in humans highly exposed to certain PCBs/PCDFs. Environ. Sci. Technol. (2006) 40(19):6176–6180.[Medline]
Maruyama W, Aoki Y. Estimated cancer risk of dioxins to humans using a bioassay and physiologically based toxicokinetic model. Toxicol. Appl. Pharmacol. (2006) 214:188–198.[CrossRef][Web of Science][Medline]
Maruyama W, Yoshida K, Tanaka T, Nakanishi J. Possible range of dioxin concentration in human tissues: Simulation with a physiologically based model. J. Toxicol. Environ. Health A (2002) 65:2053–2073.[CrossRef][Web of Science][Medline]
Masuda Y. Fate of PCDF/PCB congeners and change of clinical symptoms in patients with Yusho PCB poisoning for 30 years. Chemosphere (2001) 43:925–930.[Medline]
Meek ME, Bucher JR, Cohen SM, Dellarco V, Hill RN, Lehman-McKeeman LD, Longfellow DG, Pastoor T, Seed J, Patton DE. A framework for human relevance analysis of information on carcinogenic modes of action. Crit. Rev. Toxicol. (2003) 33:591–653.[Web of Science][Medline]
Moser GA, McLachlan MS. The influence of dietary concentration on the absorption and excretion of persistent lipophilic organic pollutants in the human intestinal tract. Chemosphere (2001) 45:201–211.[Medline]
National Academy of Sciences (NAS). Health Risks from Dioxin and Related Compounds: Evaluation of the EPA Reassessment (2006) Available at: http://www.nap.edu/catalog/11688.html. Accessed January 23, 2008.
National Toxicology Program (NTP). NTP toxicology and carcinogenesis studies of 2,3,4,7,8-pentachlorodibenzofuran (PeCDF) (CAS No. 57117-31-4) in female Harlan Sprague-Dawley rats (gavage studies). Natl. Toxicol. Program Tech. Rep. Ser. (2006a) 525:1–198.[Medline]
National Toxicology Program (NTP). NTP toxicology and carcinogenesis studies of 2,3,7,8-tetrachlorodibenzodioxin (TCDD) (CAS No. 1746-01-6) in female Harlan Sprague-Dawley rats (gavage studies). Natl. Toxicol. Program Tech. Rep. Ser. (2006b) NIH Publication No. 04-4455. Peer Reviewed February 17, 2004.
Nohara K, Ao K, Miyamoto Y, Ito T, Suzuki T, Toyoshiba H, Tohyama C. Comparison of the 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)-induced CYP1A1 gene expression profile in lymphocytes from mice, rats, and humans: Most potent induction in humans. Toxicology (2006) 225(2–3):204–213.[CrossRef][Web of Science][Medline]
Pitot HC, Goldsworthy TL, Moran S, Kennan W, Glauert HP, Maronpot RR, Campbell HA. A method to quantitate the relative initiating and promoting potencies of hepatocarcinogenic agents in their dose-response relationships to altered hepatic foci. Carcinogenesis (1987) 8(10):1491–1499.
Popp JA, Crouch E, McConnell EE. A weight-of-evidence analysis of the cancer dose-response characteristics of 2,3,7,8-tetrachlorodibenzodioxin (TCDD). Toxicol. Sci. (2006) 89:361–369.
Randerath E, Randerath K, Reddy R, Narasimhan TR, Wang X, Safe S. Effects of polychlorinated dibenzofurans on compounds in hepatic DNA of female Sprague-Dawley rats: Structure dependence and mechanistic considerations. Chem. Biol. Interact. (1993) 88(2–3):175–190.[CrossRef][Web of Science][Medline]
Randerath K, Putman KL, Randerath E, Mason G, Kelley M, Safe S. Organ-specific effects of long term feeding of 2,3,7,8-tetrachlorodibenzo-p-dioxin and 1,2,3,7,8-pentachlorodibenzo-p-dioxin on I-compounds in hepatic and renal DNA of female Sprague-Dawley rats. Carcinogenesis (1988) 9(12):2285–2289.
Randerath K, Putman KL, Randerath E, Zacharewski T, Harris M, Safe S. Effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin on I-compounds in hepatic DNA of Sprague-Dawley rats: Sex-specific effects and structure-activity relationships. Toxicol. Appl. Pharmacol. (1990) 103(2):271–280.[CrossRef][Web of Science][Medline]
Roberts ES, Hopkins NE, Foroozesh M, Alworth WL, Halpert JR, Hollenberg PF. Inactivation of cytochrome P450s 2B1, 2B4, 2B6, and 2B11 by arylalkynes. Drug Metab. Dispos. (1997) 25(11):1242–1248.
Roos PH, Belik R, Follmann W, Degen GH, Knopf HJ, Bolt HM, Golka K. Expression of cytochrome P450 enzymes CYP1A1, CYP1B1, CYP2E1 and CYP4B1 in cultured transitional cells from specimens of the human urinary tract and from urinary sediments. Arch. Toxicol. (2006) 80(1):45–52.[CrossRef][Web of Science][Medline]
Ruby MV, Casteel SW, Evans TJ, Fehling KA, Paustenbach DJ, Budinsky RA, Giesy JP, Aylward LL, Landenberger BD. Rapid communication: Background concentrations of dioxins, furans and PCBs in Sprague-Dawley rats and Juvenile Swine. J. Toxicol. Environ. Health A (2004) 67:845–850.[CrossRef][Web of Science][Medline]
Ryan JJ, Gasiewicz TA, Brown JF Jr. Human body burden of polychlorinated dibenzofurans associated with toxicity based on the Yusho and Yucheng incidents. Fundam. Appl. Toxicol. (1990) 15:722–731.[CrossRef][Web of Science][Medline]
Ryan JJ, Levesque D, Panopio LG, Sun WF, Masuda Y, Kuroki H. Elimination of polychlorinated dibenzofurans (PCDFs) and polychlorinated biphenyls (PCBs) from human blood in the Yusho and Yu-Cheng rice oil poisonings. Arch. Environ. Contam. Toxicol. (1993) 24(4):504–512.[CrossRef][Web of Science][Medline]
Schrenk D, Lipp H-P, Wiesmuller T, Hagenmaier H, Bock KW. Assessment of biological activities of mixtures of polychlorinated dibenzo-p-dioxins: Comparison between defined mixtures and their constituents. Arch. Toxicol. (1991) 65:114–118.[CrossRef][Web of Science][Medline]
Schrenk D, Schmitz HJ, Bohnenberger S, Wagner B, Worner W. Tumor promoters as inhibitors of apoptosis in rat hepatocytes. Toxicol. Lett. (2004) 149(1–3):43–50.[CrossRef][Web of Science][Medline]
Schrenk D, Stilven T, Gohl G, Viebahn R, Bock KW. Induction of CYP1A and glutathione-S-transferase by 2,3,7,8-tetrachlorodibenzo-p-dioxin in human hepatocyte cultures. Carcinogenesis (1995) 16(4):943–946.
Schwarz M, Appel KE. Carcinogenic risks of dioxin: Mechanistic considerations. Regul. Toxicol. Pharmacol. (2005) 43:19–34.[CrossRef][Web of Science][Medline]
Seed J, Carney EW, Corley RA, Crofton KM, DeSesso JM, Foster PM, Kavlock R, Kimmel G, Klaunig J, Meek ME, et al. Overview: Using mode of action and life stage information to evaluate the human relevance of animal toxicity data. Crit. Rev. Toxicol. (2005) 35:664–672.[Medline]
Shu HP, Paustenbach DJ, Murray FJ. A critical evaluation of the use of mutagenesis, carcinogenesis, and tumor promotion data in a cancer risk assessment of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Regul. Toxicol. Pharmacol. (1987) 7(1):57–88.[CrossRef][Web of Science][Medline]
Silkworth JB, Koganti A, Illouz K, Possolo A, Zhao M, Hamilton SB. Comparison of TCDD and PCB CYP1A induction sensitivities in fresh hepatocytes from human donors, Sprague-Dawley rats, and rhesus monkeys and HepG2 cells. Toxicol. Sci. (2005) 87(2):508–519.
Slikker W Jr, Andersen ME, Bogdanffy MS, Bus JS, Cohen SD, Conolly RB, David RM, Doerrer NG, Dorman DC, Gaylor DW, et al. Dose-dependent transitions in mechanisms of toxicity. Toxicol. Appl. Pharmacol. (2004a) 201(3):203–225.[CrossRef][Web of Science][Medline]
Slikker W Jr, Andersen ME, Bogdanffy MS, Bus JS, Cohen SD, Conolly RB, David RM, Doerrer NG, Dorman DC, Gaylor DW, et al. Dose-dependent transitions in mechanisms of toxicity: Case studies. Toxicol. Appl. Pharmacol. (2004b) 201(3):226–294.[CrossRef][Web of Science][Medline]
Sonich-Mullin C, Fielder R, Wiltse J, Baetcke K, Dempsey J, Fenner-Crisp P, Grant D, Hartley M, Knaap A, Kroese D, et al. IPCS conceptual framework for evaluating a mode of action for chemical carcinogenesis. Regul. Toxicol. Pharmacol. (2001) 34:146–152.[CrossRef][Web of Science][Medline]
Stinchcombe S, Buchmann A, Bock KW, Schwarz M. Inhibition of apoptosis during 2,3,7,8-tetrachlorodibenzo-p-dioxin-mediated tumour promotion in rat liver. Carcinogenesis (1995) 16(6):1271–1275.
Teeguarden JG, Dragan YP, Singh J, Vaughan J, Xu YH, Goldsworthy T, Pitot HC. Quantitative analysis of dose- and time-dependent promotion of four phenotypes of altered hepatic foci by 2,3,7,8-tetrachlorodibenzo-p-dioxin in female Sprague-Dawley rats. Toxicol. Sci. (1999) 51:211–223.
ten Tusscher GW, Guchelaar HJ, Koch J, Ilsen A, Vulsma T, Westra M, van der Slikke JW, Olie K, Koppe JG. Perinatal dioxin exposure, cytochrome P-450 activity, liver functions and thyroid hormones at follow-up after 7-12 years. Chemosphere (2008) 70(10):1865–1872.[Medline]
Tijet N, Boutros PC, Moffat ID, Okey AB, Tuomisto J, Pohjanvirta R. Aryl hydrocarbon receptor regulates distinct dioxin-dependent and dioxin-independent gene batteries. Mol. Pharmacol. (2006) 69(1):140–153.
Toyoshiba H, Walker NJ, Bailer AJ, Portier CJ. Evaluation of toxic equivalency factors for induction of cytochromes P450 CYP1A1 and CYP1A2 enzyme activity by dioxin-like compounds. Toxicol. Appl. Pharmacol. (2004) 194(2):156–168.[CrossRef][Web of Science][Medline]
Turteltaub KW, Felton JS, Gledhill BL, Vogel JS, Southon JR, Caffee MW, Finkel RC, Nelson DE, Proctor ID, Davis JC. Accelerator mass spectrometry in biomedical dosimetry: Relationship between low-level exposure and covalent binding of heterocyclic amine carcinogens to DNA. Proc. Natl. Acad. Sci. U.S.A. (1990) 87(14):5288–5292.
University of Michigan Dioxin Exposure Study (UMDES) (2006) Available at: http://www.sph.umich.edu/dioxin/index.html. Accessed January 23, 2008.
U.S. Environmental Protection Agency (USEPA). Interim Procedures for Estimating Risks Associated with Exposures to Mixtures of Chlorinated Dibenzo-p-Dioxins and -Dibenzofurans (CDDs and CDFs) and 1989 Update (1989) Risk Assessment Forum, Washington, DC. EPA/625/3–89.016.
USEPA. Exposure and Human Health Reassessment of 2,3,7,8-Tetrachlorodibenzo-p-Dioxin (TCDD) and Related Compounds (2003) NAS Review Draft. EPA/600/P-00/001Cb.
USEPA. Guidelines for Carcinogen Risk Assessment (2005) EPA/630/P-03/001F.
USEPA. Approaches for the Application of Physiologically Based Pharmacokinetic (PBPK) Models and Supporting Data in Risk Assessment (2006) EPA/600/R-05/043F August 2006. Available at: http://cfpub.epa.gov/ncea/CFM/nceaQFind.cfm?keyword=PBPK%20Models. Accessed January 23, 2008.
U.S. Environmental Protection Agency Science Advisory Board (USEPA SAB). Dioxin Reassessment-An SAB Review of the Office of Research and Development's Reassessment of Dioxin. Review of the Revised Sections (Dose Response Modeling, Integrated Summary, Risk Characterization, and Toxicity Equivalency Factors) of the EPA's Reassessment of Dioxin by the Dioxin Reassessment Review Subcommittee of the EPA Science Advisory Board (SAB) (2001) EPA-SAB-EC-01–006. Science Advisory Board, Washington, DC, May 2001 [online]. Available: http://www.epa.gov/ttn/atw/ec01006.pdf. Accessed January 23, 2008.
Van den Berg M, Birnbaum LS, Denison M, De VM, Farland W, Feeley M, Fiedler H, Hakansson H, Hanberg A, Haws L, et al. The 2005 World Health Organization reevaluation of human and Mammalian toxic equivalency factors for dioxins and dioxin-like compounds. Toxicol. Sci. (2006) 93:223–241.
van Duursen MB, Sanderson JT, Van den BM. Cytochrome P450 1A1 and 1B1 in human blood lymphocytes are not suitable as biomarkers of exposure to dioxin-like compounds: Polymorphisms and interindividual variation in expression and inducibility. Toxicol. Sci. (2005) 85(1):703–712.
Viluksela M, Bager Y, Tuomisto JT, Scheu G, Unkila M, Pohjanvirta R, Flodstrom S, Kosma VM, Maki-Paakkanen J, Vartiainen T, et al. Liver tumor-promoting activity of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in TCDD-sensitive and TCDD-resistant rat strains. Cancer Res. (2000) 60(24):6911–6920.
Waern F, Flodstrom S, Busk L, Kronevi T, Nordgren I, Ahlborg UG. Relative liver tumour promoting activity and toxicity of some polychlorinated dibenzo-p-dioxin- and dibenzofuran-congeners in female Sprague-Dawley rats. Pharmacol. Toxicol. (1991) 69:450–458.[Web of Science][Medline]
Walker NJ. Unraveling the complexities of the mechanism of action of dioxins. Toxicol. Sci. (2007) 95:297–299.
Watson MA, Devereux TR, Malarkey DE, Anderson MW, Maronpot RR. H-ras oncogene mutation spectra in B6C3F1 and C57BL/6 mouse liver tumors provide evidence for TCDD promotion of spontaneous and vinyl carbamate-initiated liver cells. Carcinogenesis (1995) 16(8):1705–1710.
World Health Organization (WHO). Assessment of the health risk of dioxins: Re-evaluation of the Tolerable Daily Intake (TDI). (1998) Geneva, Switzerland: WHO consultation.
World Health Organization (WHO). Consultation on assessment of the health risk of dioxins; re-evaluation of the tolerable daily intake (TDI): Executive summary. Food Additiv. Contam. (2000) 17:223–240.
Xu L, Li AP, Kaminski DL, Ruh MF. 2,3,7,8-Tetrachlorodibenzo-p-dioxin induction of cytochrome P4501A in cultured rat and human hepatocytes. Chem. Biol. Interact. (2000) 124(3):173–189.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. Simon, L. L. Aylward, C. R. Kirman, J. C. Rowlands, and R. A. Budinsky Estimates of Cancer Potency of 2,3,7,8-Tetrachlorodibenzo(p)dioxin Using Linear and Nonlinear Dose-Response Modeling and Toxicokinetics Toxicol. Sci., December 1, 2009; 112(2): 490 - 506. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||