Sodium perchlorate

Endpoint summary

Description of key information

Text was truncated. The full text is available in the migration report.

Key value for chemical safety assessment

Bioaccumulation potential:

no bioaccumulation potential

Additional information

This discussion compiles:

- The Toxicokinetics section from ATSDR's Toxicological profile for perchlorates (2008)

- Experimental data on Sodium Chlorate dermal penetration, and the justification for their read-across to sodium perchlorate.

Absorption:

- inhalative: possible (due to aqueous solubility) and has occurred in human studies; extent depends on particle size

- oral: very rapid (excreted from 10 min) and near complete (up to 95%) in humans and animals.

- dermal: expected to be low (ion, poor diffusion across lipid membranes)

The dermal absorption of Sodium Chlorate was very low with at most 1.85% (total extent incl. amount in stratum corneum) and 0.446 µg/cm2/hour (rate) over a 24-hour in vitro exposure of human skin, with total recovery (99-101%). This value is extrapolated to Sodium Perchlorate, based on the structural closeness of both anions (chlorate: ClO3-; perchlorate: ClO4-) and the fact that the toxicity of Sodium Perchlorate is driven by the perchlorate anion. This extrapolation is supported by the following table summarizing differential features of (Sodium) Chlorate and (Sodium) Perchlorate and their impact on dermal absorption:

Determinants of dermal penetration: differential analysis between Sodium Chlorate and Sodium perchlorate (based on ECHA's Guidance on information requirements and chemical safety assessment, Table R.7.12-3, 2008)

Determinant of dermal penetration		Influence on penetration: favors (+) or limits (-)	Differential influence: SP when compared with SC:higher (+), lower (-) or similar (=) penetration
Physical state	- Liquid / solution	+ (quicker)	- (SP: solid / SC: solution)
	- Solid	- (slower)
Molecular weight below 100		+ (quicker)	- to = (Perchlorate= 99.5 / Chlorate= 83.5; SP= 122.4 / SC = 106.4)
Structural alerts for skin binding*		- (bound = non absorbable)	= (neither SP nor SC present such alerts)
Water solubility		+ above 0.1 g/L	= (20°C: SP= 2090g/L vs. SC= 696-736 g/L***)
Calculated log Kow**		- below -1 (low solubility in stratum corneum)	= (SP not applicable, inorganic / SC<-2.9***)
Vapour pressure		- above 100 Pa (volatilizes)	= (SP: not applicable, solid / SC: <3.5 10-5 Pa***)
Surface active properties		+ below 10 mN/m	= (neither SP nor SC are surface active)
Skin irritation potential		+ (damages skin)	- (SP: non-irritant / SC skin irritant***)

SC: sodium chlorate as tested in the dermal penetration study under 7.1.2: formulations in water

SP: Sodium Perchlorate as in actual use i.e. solid particles (no formulation is used)

*: metal ions, acrylates, quaternary ammonium ions, heterocyclic ammonium ions, sulphonium salts, quinines, dialkyl sulphides, acid chlorides, halotriazines, dinitro or trinitro benzenes

**: calculated as log (Solubility in n-octanol in g/L/Solubility in water in g/L)

***: taken from EC, Draft Assessment Report (DAR) - Chlorate, 2008

Based on the above table, the dermal penetration of Perchlorate and Chlorate anions is essentially similar and in both cases very low shows a negligible ability to cross stratum corneum. Furthermore, Perchlorate is expected to penetrate less than the tested Chlorate (solution) due to exposure to solid particles, slightly higher molecular weight, and absence of skin irritant potential.

Distribution:

- Binds to serum proteins (albumin; weakly to transferrin)

- Distributes broadly

- Concentrates in thyroids (5-10x serum levels) and to a much lower extent salivary gland and skin (both 1-2x) due to active saturable transport mechanism (NIS protein).

- No accumulation: thyroid elimination t1/2 of 10-20h in rat.

- Secreted in the gastric lumen

- Transferred to milk and across the placenta (by possibly saturable mechanisms)

Metabolism:

- no cleavage of chloride or oxygen atoms in radiolabel studies: excreted unchanged

- no covalent binding to thyroid proteins

Elimination/excretion:

- mostly urinary excretion

- milk excretion into milk

- rapid and near complete serum depuration: t1/2 of 8–12 hours in humans and 10–20 hours in rats, possibly due to interspecies differences in binding proteins and/or gastrointestinal transfer by NIS protein

- does not bioaccumulate along prolonged exposure

- probably first-order kinetics in urine

PBPK models for risk assessment:

- models have been developed based on experimental data in adult, pregnant and lactating rats and in human volunteers exposed to perchlorate

- they were validated by the good correlation between model predictions and experimental measures, for use in rat and humans

- they enable to refine the interspecies kinetic assessment factor (quantified species difference in level of iodine uptake inhibition for a given external dose in mg/kg/day)

- they do not address the inter-species dynamic assessment factor (relation between a given thyroid exposure level and the hormonal and histological adverse effects)

Discussion on bioaccumulation potential result:

Toxicokinetics section (3.4) from ATSDR's Toxicological profile for perchlorates (2008), full document attached in this section.

3.4.1 Absorption

3.4.1.1 Inhalation Exposure

No studies were found regarding quantitative absorption of perchlorate after inhalation exposure. Occupational studies have measured urinary perchlorate in workers, suggesting that pulmonary absorption may occur (Lamm et al. 1999), although swallowing of particles may have also occurred. Under normal ambient temperatures, the vapor pressure of a perchlorate salt solution is expected to be low, which would reduce the likelihood of exposure to perchlorate fumes or vapors from that source. However, if perchlorate particles were suspended in air, absorption by inhalation would be possible, depending on the particle size. It is also possible that a portion of perchlorate particles suspended in the air could be swallowed and absorbed orally. Given the aqueous solubility of perchlorate salts, it is likely that small particles reaching the alveoli would dissolve and readily enter the systemic circulation.

3.4.1.2 Oral Exposure

Perchlorate has been shown, in both human and animal studies, to be readily absorbed after oral exposure. In human subjects who ingested 10 mg/day perchlorate as potassium perchlorate in drinking water for 14 days (0.14 mg/kg/day), urinary excretion rate of perchlorate was 77% of the dose/day, after 7 days of exposure, indicating that at least 77% of the ingested dosage had been absorbed (Lawrence et al. 2000). Evidence for rapid absorption in humans is provided by studies of elimination patterns. Anbar et al. (1959) detected potassium perchlorate in urine samples collected from four subjects 3 hours after ingestion of 200 mg perchlorate. Durand (1938) gave sodium perchlorate in a single oral dose (784 mg perchlorate per person) to two individuals and found perchlorate in the urine as early as 10 minutes after ingestion. Approximately 30% of the ingested dose had been eliminated in the urine within 3 hours after the dose, and 95% was eliminated within 48 hours. In a study of 13 subjects given 0.5 or 3 mg perchlorate/day for 6 months, serum perchlorate increased from undetected at baseline to an average of

24.5 μg/L in the low-dose group and 77.9 μg/L in the high-dose group over the 6 months (Braverman et al. 2006). The investigators estimated that approximately 65–70% of the daily dose was excreted during a 24-hour period. These results suggest rapid and near complete absorption of perchlorate through the digestive system.

Selivanova et al. (1986) examined the absorption of ammonium perchlorate in rats, rabbits, and calves after a single oral dose (2, 20, 200, or 600 mg perchlorate/kg). In rats, a maximum concentration of perchlorate in blood was noted between 30 and 60 minutes after administration (suggesting entrance into the systemic circulation before 30 minutes); in cattle, the maximum blood concentration of perchlorate occurred at 5 hours. In this study, only 8.5% of the administered dose was excreted in feces, and the rest was excreted in the urine, suggesting that >90% of the administered oral dose was absorbed.

3.4.1.3 Dermal Exposure

No studies were found regarding absorption of perchlorate after dermal exposure. As a general rule, electrolytes applied from aqueous solutions do not readily penetrate the skin (Scheuplein and Bronaugh 1983). On this basis, dermal absorption of perchlorate is expected to be low.

3.4.2 Distribution

Perchlorate binds to bovine and human serum albumin (Carr 1952; Scatchard and Black 1949). Perchlorate binds only weakly to either of the two binding sites of transferrin (association constants 7 and) (Harris et al. 1998).

Studies conducted in rabbits and rats indicate that perchlorate concentrations in most soft tissues (e.g., kidney, liver, skeletal muscle) are similar to the serum concentrations; tissue:serum concentration ratios >1 have been found in thyroid (5–10) and skin (1–2) (Durand 1938; Yu et al. 2002). Accumulation of perchlorate in the thyroid occurs by a saturable, active transport process (see Section 3.5.1). As a result, thyroid serum concentrations and the amount of perchlorate in the thyroid as a fraction of the absorbed dose decrease with increasing dose (Chow and Woodbury 1970). Elimination of perchlorate from the thyroid gland is relatively rapid, with half-times in rats estimated to be approximately 10–20 hours (Fisher et al. 2000; Goldman and Stanbury 1973; Yu et al. 2002).

Studies conducted in rats administered intravenous injections of perchlorate indicate that perchlorate is secreted into the gastric lumen (Yu et al. 2002). Perchlorate secreted into the gastric lumen may be absorbed in the small intestine.

3.4.2.1 Inhalation Exposure

No studies were found in humans or in animals regarding distribution of perchlorate after inhalation exposure.

3.4.2.2 Oral Exposure

In a survey of 36 healthy lactating volunteers, perchlorate was detected in breast milk at a mean concentration of 10.5 μg/L (range, 0.6–92. μg/L) (Kirk et al. 2005). Exposure of the lactating women was presumed to have occurred mainly from perchlorate in food and drinking water. No correlation was apparent between the concentration of perchlorate in the breast milk and the water that the respective mothers consumed. Serial collection of breast milk from 10 lactating women over a 3 -day period revealed that the concentrations of perchlorate, iodide, and thiocyanate varied significantly over time (Kirk et al. 2007). For perchlorate, the range, mean and median in 147 samples were 0.5–39.5, 5.8, and 4.0 μg/L, respectively. A study of women from three different cities in Chile also detected perchlorate in breast milk at mean concentrations ranging from 17.7 to 95.6 μg/L (Téllez et al. 2005). This study also found no significant correlations between breast milk perchlorate and either urine perchlorate or breast milk iodine concentrations. A study of 57 lactating women in Boston reported a median concentration of perchlorate in milk of 9.1 μg/L (range 1.3–411 μg/L) (Pearce et al. 2007).

Perchlorate also has been detected in dairy milk. A survey of 12 U.S. states showed a mean milk perchlorate level of 5.81 μg/L in 125 samples (FDA 2007a, 2007b), which was lower than a reported mean of 9.39 μg/L for Japanese samples (Dyke et al. 2007). The recent Total Diet Study (TDS) study conducted by the FDA reported a mean concentration of perchlorate of 7 μg/L in eight samples of milk (Murray et al. 2008).

Studies conducted in rabbits and rats indicate that perchlorate concentrations in most soft tissues (e.g., kidney, liver, skeletal muscle) are similar to the serum concentrations; tissue:serum concentration ratios >1 have been found in thyroid (5–10) and skin (1–2) (Durand 1938; Yu et al. 2002). Accumulation of perchlorate in the thyroid occurs by a saturable, active transport process (see Section 3.5.1).

Perchlorate has been shown to cross the placenta of rats. In rats exposed to perchlorate in drinking water, fetal:maternal serum concentration ratios were approximately 1 when the maternal dosage was 1 mg/kg/day or lower, and were <1 when the maternal dosage was 10 mg/kg/day, suggesting the possibility of a dose-dependent limitation in the capacity of transplacental transfer (Clewell et al. 2003a).

3.4.2.3 Dermal Exposure

No studies were found regarding distribution of perchlorate after dermal exposure.

3.4.2.4 Other Routes of Exposure

Several studies have examined the distribution of perchlorate in animals after intravenous, intramuscular, or peritoneal injection (Anbar et al. 1959; Chow and Woodbury 1970; Chow et al. 1969; Durand 1938; Goldman and Stanbury 1973; Yu et al. 2002). These studies have shown that absorbed perchlorate, regardless of the route of exposure, will distribute to soft tissues, including adrenal, brain, kidney, liver, mammary gland, skeletal muscle, spleen, testes, and thyroid. The highest concentrations occur in the thyroid, where tissue:serum concentration ratios of 5–10 have been observed (Chow and Woodbury 1970). The elimination half-time for the thyroid was estimated in rats to be approximately 10–20 hours (Fisher et al. 2000; Goldman and Stanbury 1973; Yu et al. 2002).

Other tissues that appear to concentrate perchlorate are the salivary gland and skin, although not to the same degree as the thyroid (Anbar et al. 1959; Lazarus et al. 1974; Yu et al. 2002). Tissue:blood concentration ratios of 1.5–2 have been observed for the salivary gland (Anbar et al. 1959) and 1–2 for the skin (Yu et al. 2002).

3.4.3 Metabolism

There is no evidence that perchlorate is metabolized in the body. Anbar et al. (1959) assayed for potential metabolites of potassium perchlorate (radiolabeled with 36Cl and 18O4) in the urine of patients 3 hours after a single oral dose (200 mg perchlorate per person). They did not detect any isotopic exchange of the oxygen atoms in excreted perchlorate; furthermore, although they found that 1–3% of the excreted 36Cl was chloride ion, this value was within experimental error. They concluded that the perchlorate excreted after 3 hours was unmodified. There has been no investigation as to whether perchlorate that is eliminated at later time points would exhibit the same isotopic pattern.

Goldman and Stanbury (1973) found that perchlorate reached a maximum concentration (>3% of the administered dose/g tissue) in the thyroid gland of rats 4 hours after an intraperitoneal injection of radiolabeled potassium perchlorate (K 36ClO4; 18 or 24 mg perchlorate/kg). However, trichloroacetic acid precipitates of thyroid homogenates contained only background levels of radioactivity, indicating that perchlorate is not covalently bound to thyroid protein.

3.4.4 Elimination and Excretion

The few studies of the elimination and excretion of perchlorate, described in the sections that follow, suggest that it is rapidly eliminated from the body through the urinary tract. Similar results have been obtained after oral exposure or after intravenous or intraperitoneal injection; the specific cation appears not to influence the pattern of excretion.

3.4.4.1 Inhalation Exposure

A study in two workers occupationally exposed to perchlorate found that the urinary perchlorate concentration increased over 3 days of perchlorate exposure, but there was a decrease between the 12-hour work shifts (Lamm et al. 1999). Excretion after the last exposure appeared to follow a first-order kinetics pattern, particularly when the urinary perchlorate concentration was between 0.1 and 10 mg/L. The average elimination half-life for the two workers was approximately 8 hours. No information was located regarding excretion of perchlorate in animals following inhalation exposure.

3.4.4.2 Oral Exposure

In adult human subjects who ingested potassium perchlorate in drinking water (0.14 mg/kg/day) for 14 days, urinary excretion rate of perchlorate was 77% of the dose/day after 7 days of exposure, indicating that urine is the main excretory pathway for absorbed perchlorate (Lawrence et al. 2000). The urinary excretion rate of perchlorate returned to control levels (<0.5 mg/day) within 14 days after exposure to perchlorate was terminated (Lawrence et al. 2000). Perchlorate was detected in the urine of two adults at 10 minutes after a single oral dose of sodium perchlorate (784 mg perchlorate per person); urinary excretion as a percentage of the dose was 30% at 3 hours, 50% in at 5 hours, 85% at 24 hours, and 95% at 48 hours (Durand 1938). This suggests an excretion half-time of approximately 12 hours. The latter estimate is consistent with the elimination kinetics of perchlorate from serum. The elimination halftime for perchlorate in serum was estimated to be approximately 8 hours in adult human subjects who ingested potassium perchlorate in drinking water (0.5 mg/kg/day) for 14 days (Greer et al. 2002). In another study in adult humans, it was estimated that approximately 65–70% of a daily dose of 0.5–3 mg perchlorate/day was excreted over a 24-hour period (Braverman et al. 2006). Thus, in humans, perchlorate is rapidly eliminated and would not be expected to accumulate in the body with prolonged exposure. Based on an elimination half-time of approximately 8–12 hours, a steady state would be achieved within 3–4 days of continuous exposure. The detection of perchlorate in breast milk from lactating women (Kirk et al. 2005; Pearce et al. 2007; Téllez et al. 2005) also indicates breast milk as an excretion route in humans.

Studies conducted in a variety of experimental animals, including rats, rabbits, and calves, have shown that absorbed perchlorate is rapidly and nearly completely excreted in the urine (Fisher et al. 2000; Selivanova et al. 1986; Yu et al. 2002).

Studies conducted in rats have shown that perchlorate is excreted in mammary milk (Clewell et al. 2003b). Perchlorate has also been detected in dairy milk (Howard et al. 1996; Kirk et al. 2005).

3.4.4.3 Dermal Exposure

No studies were found regarding elimination or excretion of perchlorates after dermal exposure.

3.4.4.4 Other Routes of Exposure

Studies in which rats received intravenous or intraperitoneal injections of perchlorate provide additional support for the rapid excretion of perchlorate in urine. Rats that received a single intravenous injection of 0.01, 0.1, 1.0, or 3.0 mg/kg perchlorate (as ammonium perchlorate) excreted 85, 86, 80, or 79% of the administered dose, respectively, in urine (Fisher et al. 2000). The elimination half-time for intravenously injected perchlorate (approximately 0.04 mg, 0.18–0.25 mg/kg, as potassium perchlorate) from serum, and the urinary excretion half-time were estimated in rats to be approximately 20 hours (Goldman and Stanbury 1973). Similarly, rats injected with sodium perchlorate (2, 8, or 49 mg perchlorate/kg) excreted 50% of the administered dose in urine during the first 6 hours and had excreted 93–97% of the dose by 60 hours (Eichler and Hackenthal 1962); in this study, higher doses of perchlorate were eliminated at a faster rate than lower doses. Similar results were obtained in rats that received a single intravenous dose of 3.3 mg/kg perchlorate as ammonium perchlorate; urinary excretion of perchlorate was essentially complete within 12 hours (Yu et al. 2002). Possible contributors to the relatively longer elimination half-life of perchlorate in rats than in humans include differences in serum protein binding or perhaps the NIS protein in the gastrointestinal tract may sequester perchlorate temporarily to a greater degree in the rat than human.

3.4.5 Physiologically Based Pharmacokinetic (PBPK)/Pharmacodynamic (PD) Models

Physiologically based pharmacokinetic (PBPK) models use mathematical descriptions of the uptake and disposition of chemical substances to quantitatively describe the relationships among critical biological processes (Krishnan et al. 1994). PBPK models are also called biologically based tissue dosimetry models. PBPK models are increasingly used in risk assessments, primarily to predict the concentration of potentially toxic moieties of a chemical that will be delivered to any given target tissue following various combinations of route, dose level, and test species (Clewell and Andersen 1985). Physiologically based pharmacodynamic (PBPD) models use mathematical descriptions of the dose-response function to quantitatively describe the relationship between target tissue dose and toxic end points.

PBPK/PD models refine our understanding of complex quantitative dose behaviors by helping to delineate and characterize the relationships between: (1) the external/exposure concentration and target tissue dose of the toxic moiety, and (2) the target tissue dose and observed responses (Andersen and Krishnan 1994; Andersen et al. 1987). These models are biologically and mechanistically based and can be used to extrapolate the pharmacokinetic behavior of chemical substances from high to low dose, from route to route, between species, and between subpopulations within a species. The biological basis of PBPK models results in more meaningful extrapolations than those generated with the more conventional use of uncertainty factors.

The PBPK model for a chemical substance is developed in four interconnected steps: (1) model representation, (2) model parameterization, (3) model simulation, and (4) model validation (Krishnan and Andersen 1994). In the early 1990s, validated PBPK models were developed for a number of toxicologically important chemical substances, both volatile and nonvolatile (Krishnan and Andersen 1994; Leung 1993). PBPK models for a particular substance require estimates of the chemical substance-specific physicochemical parameters, and species-specific physiological and biological parameters. The numerical estimates of these model parameters are incorporated within a set of differential and algebraic equations that describe the pharmacokinetic processes. Solving these differential and algebraic equations provides the predictions of tissue dose. Computers then provide process simulations based on these solutions.

The structure and mathematical expressions used in PBPK models significantly simplify the true complexities of biological systems. If the uptake and disposition of the chemical substance(s) are adequately described, however, this simplification is desirable because data are often unavailable for many biological processes. A simplified scheme reduces the magnitude of cumulative uncertainty. The adequacy of the model is, therefore, of great importance, and model validation is essential to the use of PBPK models in risk assessment.

PBPK models improve the pharmacokinetic extrapolations used in risk assessments that identify the maximal (i.e., the safe) levels for human exposure to chemical substances (Andersen and Krishnan 1994). PBPK models provide a scientifically sound means to predict the target tissue dose of chemicals in humans who are exposed to environmental levels (for example, levels that might occur at hazardous waste sites) based on the results of studies where doses were higher or were administered in different species. Figure 3-3 shows a conceptualized representation of a PBPK model.

If PBPK models for perchlorates exist, the overall results and individual models are discussed in this section in terms of their use in risk assessment, tissue dosimetry, and dose, route, and species extrapolations.

3.4.5.1 Description of the models

PBPK models of the kinetics of ingested or injected perchlorate in rats and humans have been developed (Fisher et al. 2000; Merrill et al. 2003, 2005). The models were developed simultaneously with models of radioiodide biokinetics. When combined, the perchlorate and radioiodide models simulate the competitive inhibition of radioiodide transport by perchlorate in thyroid and other tissues that have NIS activity. The adult rat model was extended to include pregnancy and maternal-fetal transfer of perchlorate, and lactation and maternal-pup perchlorate transfer through milk (Clewell et al. 2003a, 2003b). Corresponding human models of pregnancy, maternal-fetal transfer, and maternal-infant transfer of perchlorate were developed (Clewell et al. 2007).

The adult rat and human models have the same structure and differ only in values for physiological and some of perchlorate parameters. Both models simulate nine tissue compartments: blood, kidney, liver, skin, stomach, thyroid, fat, other slowly perfused tissues, and other richly perfused tissues. Uptakes from blood into the tissue vascular compartments are simulated as flow-limited processes. Distributions within blood, skin, stomach, and thyroid are simulated as diffusion limited processes with first-order clearance terms. Excretion is described with a first-order clearance term for transfer of perchlorate from the kidney into urine. Uptake of perchlorate into tissues that have NIS activity are simulated using a Michaelis-Menten approach with tissue- and species-specific maximum velocities and affinity constants that are conserved across tissues and species. This includes uptake of perchlorate into thyroid follicle cells. Secretion of perchlorate into the follicle lumen, thought to be mediated by the pendrin anion transporter, is simulated using a Michaelis-Menten approach. Upregulation of NIS (i.e., induction in response to TSH) is simulated by fitting increased maximum velocities of perchlorate and radioiodide transport into the thyroid gland. The model does not explicitly include TSH-dependence of NIS levels or other aspects of the metabolism of iodide within the thyroid (e.g., hormone production and secretion), and does not simulate changes in TSH levels resulting from NIS inhibition. Active transport of perchlorate into the stomach lumen and in skin is also simulated in the models.

Extensions of the adult models to simulate perchlorate (and radioiodide) kinetics during pregnancy in rats and humans include the addition of two additional compartments representing the mammary gland and placenta (Clewell et al. 2003a, 2003b, 2007). The structure of the human pregnancy and lactation models for perchlorate (which are identical to the corresponding rat models) are shown / parameter values for the perchlorate and radioiodide human and rat models are presented (see IUCLID section 13).

Uptake of perchlorate into the mammary gland tissue from the mammary tissue vascular space is simulated as a capacity-limited transport process, representing the activities of NIS and pendrin in this tissue. Uptake of perchlorate into the placenta from blood into placental blood is simulated as a diffusion-limited process with capacity-limited transport from placenta blood to placental tissue via NIS. Exchanges of perchlorate between the placenta and fetus are simulated with first-order clearance terms. The fetal model is identical in structure to the adult (nonpregnant) model, with adjustments in the physiological and perchlorate parameters to reflect the fetus, and the following exceptions: (1) fetal exposure is described as a first-order transfer from the placenta to the serum of the fetus; (2) clearance in the fetus is described as first-order loss from the fetal serum to the placenta; and (3) binding of iodine is not represented in the fetal thyroid and plasma.

The lactating rat and human models include a milk compartment in mammary tissue and a first-order clearance term for describing secretion of perchlorate from mammary tissue into milk (Clewell et al. 2003b, 2007). Transfer of perchlorate from milk to the neonate is simulated as a first-order clearance process. The neonate model is identical in structure to the adult (nonpregnant) model, with adjustments to the physiological and perchlorate parameter values to reflect the neonate (Clewell et al. 2003a). Parameter values for perchlorate and radioiodide in children were allometrically scaled from adult values.

3.4.5.2 Validation of the models

The rat adult perchlorate model has been evaluated for predicting kidney, serum, gastric lumen, tissue (including thyroid), and urine perchlorate concentrations in adult rats that received acute intravenous injection of radiolabeled perchlorate (36ClO4-), (Merrill et al. 2003; Yu et al. 2002). In general, model predictions were within 1–2 standard deviations of observed values. When the same parameter values were used to predict perchlorate concentrations in the thyroid in rats that were exposed to repeated doses of perchlorate in drinking water for 14 days, the model predicted lower levels of perchlorate in thyroid than were observed for dosages ≥3 mg/kg/day. At doses of perchlorate >1 mg/kg/day, only slight inhibition of thyroid radioiodide uptake was observed (Yu et al. 2002); presumably, a result of upregulation of NIS by TSH, whereas the model predicted greater inhibition. However, good correspondence with observations was achieved by adjusting the parameters for maximum velocity of transport of perchlorate and radioiodide into the thyroid gland. This adjustment mimics induction of NIS that occurs in response to elevations in serum TSH, which was observed in the rats exposed to perchlorate in drinking water (Uyttersprot et al. 1997; Yu et al. 2002). TSH stimulates other changes in radioiodide metabolism in the thyroid (e.g., hormone production and secretion) that are not simulated by the model.

The adult human model (nonpregnant, non-lactating) also predicted reasonably well (i.e., within 1– 2 standard deviations of observations) perchlorate concentrations in plasma and urine in subjects who received oral doses of perchlorate (Durand 1938; Eichler 1929; Greer et al. 2002; Kamm and Drescher 1973; Merrill et al. 2005). Model predictions of radioiodide in gastric juice, serum, thyroid, and urine following an intravenous dose of radioiodide also corresponded with observations made in healthy adults (Hays and Solomon 1965). Model predictions of thyroid radioiodine uptake in subjects who received oral doses of perchlorate agreed with observations when the kinetic parameters for iodide in the thyroid (i.e., maximum transport into the thyroid follicle) were adjusted to achieve good correspondence to the observations (Greer et al. 2002; Merrill et al. 2005). When the model was calibrated by adjusting the maximum transport rate for iodide into the thyroid follicle, it accurately predicted the observed time course for radioiodine uptake in a Graves’ disease patient who received a single tracer dose of radioiodine (Stanbury and Wyngaarden 1952); however, the model substantially overpredicted iodide uptake after the same patient received a dose of perchlorate. The error in predictions of the effect of perchlorate on iodide uptake may reflect humoral regulation of iodide transport and organification mechanisms or a response to perchlorate in Graves’ disease patients that is not simulated in the model.

The rat maternal/fetal model was evaluated by comparing predictions of perchlorate concentrations in maternal and fetal serum and maternal thyroid in rats exposed to perchlorate in drinking water (Clewell et al. 2001, 2003a). Model predictions agreed well (within 1–2 standard deviations of observations) with observations. Predictions of maternal and fetal radioiodine uptakes in thyroid were also in reasonable agreement with observations in rats that received single injections of iodine with or without single injections or oral gavage doses of perchlorate, or at the conclusion of 18 days of exposures to perchlorate in drinking water (Brown-Grant 1966; Clewell et al. 2001, 2003a; Sztanyik and Turai 1988).

Similar outcomes occurred in evaluations of the lactating dam/neonate model (Clewell et al. 2003b). The model accurately predicted serum and thyroid iodide concentrations in the dam and neonate following single intravenous injections of radioactive iodine, with or without concurrent injection of perchlorate, and in maternal thyroid following an 18-day exposure to perchlorate in drinking water (Clewell et al. 2003b). Model predictions of radioiodide levels in mammary gland and milk, in rats that did or did not receive single doses of perchlorate, corresponded with observations (Clewell et al. 2003b).

The human pregnancy and lactation models were evaluated by comparing model predictions of perchlorate concentrations in serum with observations made in pregnant women (15 and 33 weeks of gestation) and their infants at birth (from cord blood), and in a group of children (mean age 7.4 years) (Téllez et al. 2005). Exposures simulated in the model were the continuous perchlorate intake corresponding to the mean ±1 standard deviation drinking water exposure concentrations (114±13 ppm) for a cohort in the Téllez et al. (2005) study. Model predictions for maternal, fetal, and child serum perchlorate concentrations agreed well (within ±1 standard deviation of observed means) with observations. Téllez et al. (2005) also reported perchlorate concentrations in breast milk measured at 5– 6 weeks postpartum. Simulation of the continuous perchlorate intake corresponding to the group means (± standard deviation) of drinking water exposure concentrations (5.8±0.6 ppm or 114±13 ppm) yielded predictions of breast milk perchlorate concentrations that were within ±1 standard deviation of the observed means.

3.4.5.3 Risk assessment

The rat (Clewell et al. 2003a, 2003b; Merrill et al. 2003) and human models (Clewell et al. 2007; Merrill et al. 2005) can be used to estimate the human equivalent exposure level for perchlorate that would give rise to a given percent inhibition of thyroidal radioiodide uptake. The models do not include downstream effects on the thyroid axis, such as decreases in serum thyroid hormones. The Clewell et al. and Merrill et al. model estimates have been used to extrapolate dose-response relationships for perchlorate observed in rats to humans, and across various human lifestages (e.g., fetus, neonate, child, adult, pregnancy, lactation). External dose–internal dose relationships for various human life-stages predicted from the human models are presented (see IUCLID section 13; Clewell et al. 2007). The models predict a relatively high vulnerability of the fetus, pregnant woman, and lactating woman to perchlorate-induced thyroid iodine uptake, compared to other lifestages (i.e., greater inhibition of thyroid iodide uptake occurs in these lifestages in association with lower external doses), compared to neonates, child or nonpregnant or nonlactating adult). The potential impact of external exposures to perchlorate on inhibition on thyroidal radioiodide uptake is sensitive to assumptions about urinary clearance of perchlorate, especially in neonates and young infants. The estimates based on external exposures from consumption of drinking water are also dependent on assumptions regarding age-related changes in contribution of drinking water to liquid consumption across lifestages (e.g., milk in children).

3.4.5.4 Target tissues

Tissues simulated in the perchlorate models are shown (see IUCLID section 13). The models were designed to calculate perchlorate concentrations in serum and thyroid and inhibition of radioiodide uptake into the thyroid resulting from exposures to perchlorate for various lifestages (e.g., fetus, neonate, child, adult, pregnancy, lactation).

3.4.5.5 Species extrapolation

The models are designed for applications to rat or human dosimetry and cannot be applied to other species without modification and validation.

3.4.5.6 Interroute extrapolation

The models are designed to simulate intravenous or oral exposures to perchlorate and cannot be applied to other routes of exposure without modification and validation.