Regulatory Toxicology and Pharmacology 58 (2010) 10–17 Contents lists available at ScienceDirect Regulatory Toxicology and Pharmacology journal homepage: www.elsevier.com/locate/yrtphBiomonitoring Equivalents for triclosanKannan Krishnan a, Michelle Gagné a, Andy Nong b, Lesa L. Aylward c,*, Sean M. Hays da Université de Montréal, Département de santé environnementale et santé au travail, Montréal, QC, Canadab Health Canada, Ottawa, Ontario, Canadac Summit Toxicology, LLP, Falls Church, VA, USAd Summit Toxicology, LLP, Lyons, CO, USAa r t i c l e i n f o a b s t r a c tArticle history: Recent efforts worldwide have resulted in a growing database of measured concentrations of chemicals inReceived 26 March 2010 blood and urine samples taken from the general population. However, few tools exist to assist in theAvailable online 10 June 2010 interpretation of the measured values in a health risk context. Biomonitoring Equivalents (BEs) are deﬁned as the concentration or range of concentrations of a chemical or its metabolite(s) in a biologicalKeywords: medium (blood, urine, or other medium) consistent with an existing health-based exposure guideline,Biomonitoring Equivalents and are derived by integrating available data on pharmacokinetics with existing chemical risk assess-Triclosan ments. This study reviews available health-based exposure guidance values for triclosan based on recentRisk assessmentPharmacokinetics evaluations from the United States Environmental Protection Agency (US EPA), the European Commis- sion’s Scientiﬁc Committee on Consumer Products (EC SCCP) and the Australian National Industrial Chemicals Notiﬁcation and Assessment Scheme (NICNAS). BE values corresponding to the reference dose (RfD) or margin of safety (MOS) targets from these agencies were derived based on kinetic data (urinary excretion and plasma clearance) from human studies and measured blood concentration data in animal studies. Estimated BE values for urinary total triclosan (free plus conjugates) corresponding to the US EPA RfD and the EC-identiﬁed margin of safety target from the NOAEL are 6.4 and 2.6 mg/L, respectively (cor- responding to 8.3 and 3.3 mg/g creatinine, respectively). Plasma BE values corresponding to the US EPA, EC, and Australian NICNAS values are 0.3, 0.9, and 0.4 mg/L, respectively. These values may be used as screening tools for evaluation of population biomonitoring data for triclosan in a risk assessment context. Ó 2010 Elsevier Inc. All rights reserved.1. Introduction approach, the development of Biomonitoring Equivalents (BEs)1 has been proposed, and guidelines for the derivation and communi- Interpretation of measurements of concentrations of chemicals cation of these values have been developed (Hays et al., 2007, 2008;in samples of urine or blood from individuals in the general popu- LaKind et al., 2008).lation is hampered by the general lack of screening criteria for A Biomonitoring Equivalent (BE) is deﬁned as the concentrationevaluation of such biomonitoring data in a health risk context. or range of concentrations of chemical in a biological mediumWithout such screening criteria, biomonitoring data can only be (blood, urine, or other medium) that is consistent with an existinginterpreted in terms of exposure trends, but cannot be used to health-based exposure guidance value such as a reference doseevaluate which chemicals may be of concern in the context of cur- (RfD) or tolerable daily intake (TDI). BEs are estimated on the basisrent risk assessments. Such screening criteria would ideally be of existing chemical-speciﬁc pharmacokinetic data (animal or hu-based on robust datasets relating potential adverse effects to bio- man) and the point of departure (e.g., NOAEL, LOAEL, BMD) usedmarker concentrations in human populations (see, for example, in the derivation of exposure guidance values (Hays et al., 2008).the U.S. Centers for Disease Control and Prevention (CDC) bloodlead level of concern; see http://www.cdc.gov/nceh/lead/). How-ever, development of such epidemiologically-based screening cri- 1 Abbreviations used: BE, biomonitoring equivalent; BEPOD, biomonitoring equiva-teria is a resource and time-intensive effort. As an interim lent point of departure; BMD, benchmark dose, BW, body weight; CAS, chemical abstracts services; EC, European Commission, LOAEL, lowest observed adverse effect level; MOE, margin of exposure; MOS, margin of safety; NHANES, National Health and Nutrition Examination Survey; NICNAS, National Industrial Chemicals Notiﬁcation and Assessment Scheme; NOAEL, no observed adverse effect level; POD, point of * Corresponding author. Address: Summit Toxicology, LLP, 6343 Carolyn Drive, departure; PK, pharmacokinetic; RED, re-registration eligibility decision; RfD, refer-Falls Church, VA 22044, USA. ence dose; SCCP, scientiﬁc committee on consumer products; TDI, tolerable daily E-mail address: email@example.com (L.L. Aylward). intake; USEPA, United States Environmental Protection Agency.0273-2300/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved.doi:10.1016/j.yrtph.2010.06.004
K. Krishnan et al. / Regulatory Toxicology and Pharmacology 58 (2010) 10–17 11BEs are intended for use as screening tools to allow an assessment and intraspecies uncertainty factors of 10 each (USEPA, 2008a)of biomonitoring data to evaluate which chemicals have large, (Table 1).small, or no margins of safety compared to existing risk assess- The European Commission (2009) also recently evaluated tri-ments and exposure guidance values. This document presents der- closan. In that review, the European Commission’s Scientiﬁc Com-ivation of BEs for triclosan (Chemical Abstracts Services [CAS] mittee on Consumer Products (SCCP) identiﬁed a point ofRegistry number 3380-34-5). departure (POD) (NOAEL of 12 mg/kg-d) from a well-conducted ro- Triclosan (5-chloro-2-(2,4-dichlorophenoxy)phenol) is a lipo- dent chronic bioassay and recommended a target margin of safetyphilic (measured log n-octanol:water partition coefﬁcient = 4.8 (MOS) of 100. Based on this assessment, a target daily dose level(Ciba-Geigy (1990) cited by NICNAS (2009)), broad-spectrum anti- analogous to a reference dose or tolerable daily intake can alsomicrobial agent that has found widespread use in a variety of per- be derived (Table 1).sonal care products including toothpaste, mouthwash, bar soap, The information on mode of action in mammals and associateddeodorant, shower gel as well as skin-care and make-up products relevant dose metrics is limited for triclosan. The mechanism of(Engelhaupt, 2007; McGinnis, 2008; EC, 2009). It is also used in antibacterial action of triclosan has been reported to involve theconsumer products such as textiles, toys and plastic kitchenware inhibition of lipid synthesis by blocking the enoyl-acyl reductase(e.g., Adolfsson-Erici et al., 2002; Bhargava and Leonard, 1996; Per- enzyme (McMurry et al., 1998; Heath et al., 1999). Inhibition ofencevich et al., 2001; Yazdankhah et al., 2006). There has been in- fatty acid synthesis in parasites has also been documented (Suroliacreased focus on the occurrence of triclosan in biological matrices et al., 2001; Samuel et al., 2003). Further, it has been demonstratedof individuals without occupational exposures as well as on the that this chemical prevents bacterial cell growth and proliferationevaluation of the potential for endocrine-disrupting effects (Crof- by interfering with the formation of new cell membranes (Levyton et al., 2007; Zorrilla et al., 2009). A number of human biomon- et al., 1999). However, the relevance or ability of these mecha-itoring studies have reported the occurrence of triclosan in breast nisms leading to adverse health effects in humans has not beenmilk, urine and plasma (Calafat et al., 2008; Hovander et al., demonstrated (Sullivan et al., 2003; Sandborgh-Englund et al.,2002; Dayan, 2007; Allmyr et al., 2008; Wolff et al., 2007; Sand- 2006).borgh-Englund et al., 2006). The analyses of 2517 single spot urine In mammals, triclosan is reported to alter serum concentrationsamples as part of the National Health and Nutrition Examination of thyroxine (Crofton et al., 2007) and to interact with P450-depen-Survey (NHANES: 2003-04) indicated that three-quarters of the dent enzymes, UDP-glucuronosyltransferases and the human preg-samples contained triclosan (Calafat et al., 2008). nane X receptor (Hanioka et al., 1996; Jacobs et al., 2005; Wang et The present study focused on establishing BEs for triclosan al., 2004). The relevance for humans of these interactions andbased on the available data on point of departure (POD) and guide- toxicological endpoints identiﬁed in animal studies is not knownline values, as well as the pharmacokinetic information. (Calafat et al., 2008; USEPA, 2008a; Allmyr et al., 2009; EC, 2009). The effects identiﬁed at levels exceeding the NOAEL levels in the baboon study were relatively non-speciﬁc. At doses above the2. Available data and approach NOAEL in the baboon study, clinical signs of toxicity included vom- iting, failure to eat and diarrhea (Ciba-Geigy, 1977; USEPA, 2008a).2.1. Exposure guidance values, critical effects, and mode of action The rat NOAEL was based on hematotoxicity as well as decreases in absolute and relative spleen weights; at higher doses, mild clinical Based on our review of information from US, Canadian, Austra- chemistry and/or hematology changes, together with histopathol-lian and European sources, three recent risk assessments and eval- ogical changes in the liver were reported (EC, 2009; Ciba-Geigy,uations for triclosan were found to be available. Of these 1986; NICNAS, 2009).assessments, that of the Australian government (NICNAS, 2009) Even though there is no deﬁnitive information on the mode ofidentiﬁed a NOAEL of 41 lg/mL (corresponding to the steady-state action or relevant dose metrics for the triclosan-induced effectsplasma level of triclosan in rats administered 40 mg/kg/d) as a ba- in rats and baboons, it would appear that the parent chemical issis for estimating MOE based on measured plasma values in hu- likely to be the toxic form, given that triclosan does not undergomans. In essence, a BE value based on a target minimal margin of any oxidative metabolism or bioactivation reaction and has beenexposure of 100 was derived in the NICNAS (2009) risk assessment. documented to conjugate with UDP-glucuronic acid and sulfateHowever, we have included this derivation in this article for (DeSalva et al., 1989). Thus, plasma concentrations of triclosancompleteness. The US EPA Ofﬁce of Pesticide Programs completed would appear to be relevant to potential toxicity.a recent re-registration eligibility decision (RED) document for tri-closan. In that review, they based the derivation of a chronic refer- 2.2. Available pharmacokinetic dataence dose (RfD) on a chronic dietary study conducted in baboonswith a no-observed-adverse-effect-level of 30 mg/kg-d and a The available data on the pharmacokinetics of triclosan in ani-composite uncertainty factor of 100 comprised of interspecies mals and humans exposed by various routes and vehicles haveTable 1Health risk assessments and health-based exposure reference values for triclosan identiﬁed in the current study. Organization Study description Critical endpoint and dose Uncertainty factors Guideline valuea USEPA (2008a) Chronic toxicity study in NOAEL = 30 mg/kg based on clinical signs of 100 0.3 mg/kg-d baboons (Drake, 1976) toxicity such as vomiting, failure to eat and diarrhea – Interspecies UF: 10 – Intraspecies UF: 10 EC (2009) 2-Year chronic rat bioassay NOAEL = 12 mg/kg/d based on hematoxicity 100 - Interspecies UF: 10 - 0.12 mg/kg-d (DeSalva et al., 1989) and decrease in absolute and relative spleen weights Intraspecies UF: 10 NICNAS (2009) 2-Year chronic rat bioassay Plasma concentration associated with NOAEL: Target minimal margin of exposure 0.4 mg/L in plasma (Australia) (DeSalva et al., 1989) 41 mg/L based on liver effects on a plasma basis: 100 – Interspecies UF: 10 – Intraspecies UF: 10 a Corresponds to values calculated by dividing the point of departure (POD) by the applicable uncertainty factors.
12 K. Krishnan et al. / Regulatory Toxicology and Pharmacology 58 (2010) 10–17Table 2Summary of urinary excretion data for triclosan in baboons and humans. Species Study description Excretion rate References Human – 44–57% of the oral dose was excreted within 48 h Ciba-Geigy Limited (1976) from NICNAS (2009) Human Ten volunteers were orally exposed to a single dose (4 mg) 54% of the oral dose was excreted by 96 h Sandborgh-Englund et al. (2006) Baboons Oral administration of approximately 5 mg/kg bw 53–60% was excreted in urine 20–30% was Ciba-Geigy Limited (1977) from 14 C-triclosan to two male baboons excreted in the feces after 144 h NICNAS (2009)been summarized by the National Industrial Chemicals Notiﬁca- of action, is reﬂective of the available pharmacokinetic (PK) data,tion and Assessment Scheme (NICNAS) of Australia (2009) and EC and is based on the relationship between the biomarker and the(2009). Triclosan is extensively conjugated with UDP-glucuronic relevant internal dose metrics. Triclosan in plasma is likely to beacid and sulfate, and excreted via urine or feces. In humans, urinary a relevant dose metric for prediction of toxicity, but is only foundexcretion is the principal route of elimination; however, in the rat, in very low levels. Urinary triclosan (conjugates + free form), ontriclosan is preferentially eliminated via the fecal route. The fol- the other hand, exhibits greater uncertainty compared to plasmalowing discussion focuses on data relevant to the derivation of concentrations in its relation to tissue concentrations but it is stillBE values for triclosan. useful as a biomarker of exposure. Figs. 1, 2a and b, illustrate the DeSalva et al. (1989) summarized the pharmacokinetic data approaches used for computing the BE for urinary and plasma tri-associated with the 2-year chronic dietary bioassay in the rat. Spe- closan, for the two PODs and associated guideline values identiﬁedciﬁcally, the concentrations of triclosan in blood, liver and kidney for triclosan (Table 1).of treated animals were determined at 3, 6, 12, 18 and 24 monthsof treatment. The data indicate: (i) the attainment of steady-state 3.1. Urinary BE valuesduring chronic exposure to triclosan, (ii) the predominance of con-jugates rather than free form in the various matrices, and (iii) the The existing data on elimination kinetics of triclosan (as parentproportionality of triclosan concentration with exposure dose in compound or conjugated) suggest a simple mass-balance ap-the rat. The EC SCCP (2009) used these blood data to estimate plas- proach, with an assumption of steady-state intake and excretion,ma concentrations of total triclosan (free and conjugates) at se- for derivation of BE values for urinary triclosan (Fig. 1). In this re-lected time points during the bioassay (see Table 27 in EC (2009)). gard, the amount of triclosan excreted in urine every day will be A particularly relevant pharmacokinetic study in humans is that approximately equal to the human-equivalent amount ingestedof Sandborgh-Englund et al. (2006). In this study, ten subjects (5 times the urinary excretion fraction. The median of the estimatedmales and 5 females) swallowed a single oral dose of 4 mg triclo- fraction of oral triclosan dose excreted via urine from ten individ-san (average estimated dose of 0.06 mg/kg-d). The total triclosan uals (54% at 48 h) obtained from Sandborgh-Englund et al. (2006)plasma levels were determined at 1, 2, 3, 4, 6, 9, 24, 48, 72, 96 was combined with age-speciﬁc estimates of bodyweight and aver-and 192 h after dosing, whereas the total urinary levels were deter- age 24-h urinary volumes (or average 24-h creatinine excretion) tomined in 24-h composite samples. The maximal concentration and provide an estimate of the 24-h average urinary concentrationhalf-lives for plasma clearance and urinary elimination in each associated with a unit dose of triclosan per day (Table 3) for differ-individual were obtained. Overall, these data indicate median plas- ent age groups. No assessment values for children under age 6ma and urinary half-lives of 19 and 11 h, respectively. Median were presented due to the lack of reliable data on urinary volumecumulative urinary excretion of triclosan as conjugated species and creatinine excretion rates. Speciﬁcally, the estimated triclosanand free compound was 54% of the administered dose after 4 days urinary concentration on a volume associated with a unit dose of(ranging from 24% to 84%), with the majority of the elimination triclosan for a speciﬁc population sub-group was calculated usingoccurring in the ﬁrst two days. Free parent triclosan accounted the following formula:for less than 1% of the eliminated compound. Other pharmacoki- D Â BW Â F UEnetic data collected by Ciba Geigy in baboons and humans indi- CV ¼ ð1Þ Vcates similar behavior; these data on urinary kinetics of triclosanwere described by NICNAS and are summarized in Table 2 (Ciba-Geigy, 1986; NICNAS, 2009).2.3. Potential biomarkers Triclosan glucuronide is the major metabolite and urinaryexcretion is the principal route of elimination in humans. As such,the total triclosan in urine (conjugates + free form) has been usedas biomarker of exposure (e.g., Calafat et al., 2008; NICNAS,2009). The plasma (or blood) level of total triclosan is also a rele-vant marker, as determined in a 2-year toxicological bioassay(DeSalva et al., 1989). As with other biomonitoring efforts, collec-tion of urinary samples is less invasive and more straightforwardthan collection of blood samples, but sampling of blood or plasmamay provide a more toxicologically relevant measure of exposure.3. BE derivation The BE for triclosan should ideally correspond to a dose mea- Fig. 1. Schematic representation of the approach used for deriving BE values forsure in a biological matrix that relates most closely to the mode urinary total triclosan concentration (free plus conjugated compound).
K. Krishnan et al. / Regulatory Toxicology and Pharmacology 58 (2010) 10–17 13 administered dose, based on actual data. The urinary concentrations of triclosan associated with a unit dose of triclosan (D) for the dif- ferent age groups are presented in Table 3. Because the average con- centrations associated with a unit dose of triclosan varied little across age and gender groups, the average across all age groups was estimated and carried forward in calculations. Using the esti- mates of the 24-h average urinary concentration associated with a unit dose of triclosan, the urinary concentrations (on both a volume and creatinine-adjusted basis) for the human-equivalent POD (ori- ginal POD divided by the interspecies uncertainty factor) and guid- ance values listed in Table 1 were estimated and reported in Table 4. 3.2. Plasma BE values For the guidance values described in Table 1, different approaches were used in the derivation of BE values for plasma concentrations based on the available datasets (see Figs. 2a and b). Because measured blood concentrations and estimates of corre- sponding plasma concentrations were available from the chronicFig. 2a. Schematic representation of the approach used for deriving BE values for rat bioassay used as the basis of the EC, 2009 risk assessment,total triclosan in plasma based on the rat NOAEL (12 mg/kg-d) identiﬁed by EC we derived a BE starting directly from those measured data using(2009). the following approach (Fig. 2a): (1) Estimate the steady-state plasma concentration in the experimental animals dosed at the POD based on data from the underlying study (DeSalva et al., 1989). Based on the val- ues of blood concentration reported in Ciba-Geigy’s 2-year study in Sprague–Dawley rats (see DeSalva et al. 1989), the EC, 2009 converted the measured blood concentrations to an estimated average steady-state plasma concentration of 28.16 mg/L at the POD (Table 27 in EC, 2009). However, this value corresponds to the interim measured value for female rats (exposed at approximately 17 mg/kd-d), while the EC-identiﬁed NOAEL was set at the dose level for male rats (12 mg/kg-d). The corresponding interim estimated plasma concentration in male rats is 21.8 mg/L (Table 27, EC, 2009), and this value is used here for the BE derivation. (2) Apply the pharmacodynamic component of the interspecies uncertainty factor (100.5) to derive estimated plasma con- centrations for the human-equivalent BEPOD; and (3) Apply the intraspecies uncertainty factor to the BEPOD to derive the BE.Fig. 2b. Schematic representation of the approach used for deriving BE values fortotal triclosan in plasma based on the baboon study NOAEL (30 mg/kg-d) identiﬁed A similar approach was used for the NICNAS (2009) risk assess-by USEPA (2008a). ment, except that risk assessment was based entirely on the mea- sured plasma concentrations in rats in DeSalva et al. (1989), and explicitly speciﬁed a minimal margin of exposure 100 on a plasmawhere CV is the average urinary concentration on a volume basis of concentration basis to identify a target plasma concentration.triclosan, D is a unit dose of triclosan (1 lg/kg-d) as shown in Table For the NOAEL of 30 mg/kg-d in the baboons identiﬁed by USEPA3, BW is the bodyweight for the group, FUE is the urinary excretion (2008a) as the POD, the plasma values associated with the BEPOD andfraction, i.e., fraction of the applied dose excreted in the urine BEwerederivedasfollows(Fig.2b):(0.54), and V is the 24-h average urinary volume. Similarly, the cre-atinine-adjusted concentration associated with a unit dose of triclo- (1) Apply the interspecies uncertainty factor to identify thesan was calculated as follows: human-equivalent POD on an external dose basis. (2) Estimate the steady-state plasma concentration, CP, in mg/L D Â BW Â F UECC ¼ ð2Þ in humans associated with the human-equivalent POD (i.e., CE BEPOD) by dividing the daily dose, D, in mg/kg-d by the clear-where CC is the creatinine-adjusted 24-h urinary concentration of ance, CL (plasma clearance normalized to the fraction oftriclosan, and CE is the 24-h creatinine excretion rate. Data on uri- dose absorbed; 0.041 L/(h kg)), estimated by Sandborgh-nary volume and creatinine excretion rates were drawn from a vari- Englund et al. (2006) according to this formula:ety of studies (see footnotes to Table 3). It is relevant to note that DFUE in Eqs. (1) and (2) above refers to the fraction of the dose CP ¼ ð3Þappearing in the urine. It does not make any particular assumptions CL Â 24regarding bioavailability or absorbed fraction; rather it represents (3) Apply the intraspecies uncertainty factor to the BEPOD tothe ratio of the amount appearing in the urine in relation to the derive the BE.
14 K. Krishnan et al. / Regulatory Toxicology and Pharmacology 58 (2010) 10–17Table 3Assumptions for average bodyweight, 24-h urinary volume, and 24-h creatinine excretion rate and estimates of creatinine-adjusted and volume-based urinary concentration perunit dose of triclosan (lg/kg-d) by age group (considering a urinary excretion fraction of 54%)a. Age Group Bodyweightb (kg) Average 24 h urinary volumec (L) Triclosan urinary concentration per lg/kg-d (creatinine excretion, gd) steady-state dose (lg/L) (lg/g creatinine)e Children, 6–11 32 0.66 26.2 (0.5) (34.6) Adolescents, 11–16 57 1.65 18.7 (1.2) (25.7) Men, >16 70 1.7 22.2 (1.5) (25.2) Women,>16 55 1.6 18.6 (1.2) (24.8) Average, lg/L 21.4 (lg/g creatinine) (27.5) a Urinary excretion fraction of 54% from Sandborgh-Englund et al. (2006). b Estimated from Table 8-1 of USEPA (2008b). c Urinary volumes for children from Remer et al. (2006). Volumes for adults from Perucca et al. (2007). Adolescents were assumed to have urinary volumes similar toaverage values for adults. d Creatinine excretion for children and adolescents estimated from Remer et al. (2002); average creatinine excretion for boys and girls under age 13, 17 mg/kg BW per day;average creatinine excretion for adolescents, 22 mg/kg BW per day. Creatinine excretion for adults estimated based on equations from Mage et al. (2004) and average USheight and speciﬁed bodyweights. e Calculated using equations 1 (for volume-based values) or 2 (for creatinine-adjusted values). For example, using equation 1, for children aged 6–11, the triclosanconcentration in lg/L associated with a dose of 1 lg/kg-d at steady state would be 1 lg/kg-d * 32 kg/0.66 L/d = 26.2 lg/L. Table 4 Derivation of BE values for triclosan urinary concentration (on a volume and creatinine-adjusted basis) consistent with the guideline values derived from USEPA (2008a) and EC (2009), according to the scheme presented in Fig. 1. Reported concentrations are the sum of both free and conjugated triclosan in urine. BE derivation step USEPA (2008a) RfD EC (2009) risk assessment Species, endpoint Baboons, general toxicity Rats, hematological endpoint alterations POD (NOAEL)a (mg/kg-d) 30 12 UF, interspecies 10 10 Human-equivalent PODb (mg/kg-d) 3 1.2 BEPOD, mg/L in urine 64 26 (mg/g creatinine) (83) (33) UF, intraspecies 10 10 BE, mg/L in urine 6.4 2.6 (mg/g creatinine) (8.3) (3.3) a From Table 1. b Estimated using the increments in urinary concentrations per unit of steady-state dose reported in Table 3. The derivation and the resulting values for both guidance values 100.5 used in risk assessments (to represent PK variability), and thisare summarized in Table 5. uncertainty factor component was retained in the derivation of the urinary BE values presented here. The intersubject variability in the excretion fraction might result from variation of bioavailability,4. Discussion distribution kinetics, metabolic clearance and/or fraction elimi- nated via renal clearance (Sandborgh-Englund et al., 2006). Addi-4.1. Sources of variability and uncertainty tional sources of potential variation in measured urinary concentrations, even under conditions of exposure consistent with The urinary and plasma BE values for triclosan derived in this the RfD, include variations in hydration status and creatinineevaluation were based on the average values of input parameters; excretion rates, which could impact measured concentrations inhowever, the resulting values accounted for inter-individual vari- spot urine sample. The appropriateness of adjustment for hydra-ability by way of the use of uncertainty factors. tion status using creatinine excretion has been debated (Garde For the derivation of urinary BE values, the median urinary et al., 2004; Barr et al., 2005) because creatinine excretion alsoexcretion fraction of 54% of an oral dose of triclosan reported by can vary substantially due to variations in dietary pattern as wellSandborgh-Englund et al. (2006) was used. This value was obtained as other individual factors (gender, age, muscle mass, seasonalwith 10 study subjects aged between 26 and 42 years (5 females and daily variation, diet) (Garde et al., 2004; Barr et al., 2005).and 5 males). The authors reported that the major fraction of triclo- In the present work, BE values were estimated for triclosan onsan was excreted within the ﬁrst 24 h. The lower and upper quar- the basis of creatinine excretion as well as on the basis of urinarytile values of fraction excreted were 47% and 61% at 48 h; the volume. Samples collected for a 24-h period would be expectedindividual values of the fraction excreted, however, varied from to be inﬂuenced less than spot samples by both variations in24% to 83% after 4 days of exposure (median value = 54%) (Sand- hydration status and creatinine excretion. Even though the valueborgh-Englund et al., 2006). In other words, the ratio of the maxi- of excretion fraction used in the calculations (54%) was based onmal to the median value of excretion fraction for triclosan was less a limited dataset from a human volunteer study (Sandborgh-Engl-than a factor of 2, or well within the inter-individual variability of und et al., 2006), this value is consistent with other available
K. Krishnan et al. / Regulatory Toxicology and Pharmacology 58 (2010) 10–17 15Table 5Derivation of plasma BE values based on the USEPA (2008b) and EC (2009) risk assessments (Table 1) according to the schemes presented in Fig. 2a (for the EC and NICNAS riskassessments) and Fig. 2b (for the USEPA risk assessment). BE derivation step USEPA, 2008a,b) RfD EC (2009) risk assessment NICNAS (2009) risk assessment Species, endpoint Baboons, general toxicity Rats, hematological endpoint alterations Rat, liver toxicity POD (NOAEL), external dosea (mg/kg-d) 30 12 NA POD (NOAEL), plasma concentrationb (mg/L) – 21.8 41 UF, interspecies 10 2.5d 10e Human-equivalent POD, external dose (mg/kg-d) 3 – BE_POD, plasma concentrationc mg/L 3.0 8.7 4.1 UF, intraspecies 10 10 10e BE, plasma concentration (mg/L) 0.3 0.9 0.4 a From Table 1. b From DeSalva et al. (1989) as reported by EC (2009). c Estimated using Eq. (3) and median clearance rate measured by Sandborgh-Englund et al. (2006). d Because a relevant internal dose metric is used here, the pharmacokinetic component of the interspecies uncertainty factor is set to 1. We have retained the pharma-codynamic component of the interspecies uncertainty factor (2.5, according to WHO (1999), guidance used in EC risk assessments), e Speciﬁed value on a plasma concentration basis in the NICNAS risk assessment.information in the literature. In other human studies involving a et al. (2006), an estimated steady-state plasma concentration ofsingle oral dose ranging from 5 to 200 mg, the average cumulative 21.8 mg/L in rats was used. This plasma concentration was ob-amount of triclosan excreted in 24-h urine corresponded to about tained from EC (2009), based on the interim sacriﬁce data in male40%, with maximal percent approaching 60% in 4–5 days (DeSalva rats from the critical toxicological study. EC (2009) indicated thatet al., 1989). The inter-individual coefﬁcient of variation in urinary the plasma concentration at interim sacriﬁce was 21.8 (±9) andexcretion fraction was on the order of 30% in dermal and oral expo- 28.1 (±12.9) mg/L in male and female rats respectively, whereassure studies (Queckenberg et al., 2009), which is likely to be reﬂec- at terminal sacriﬁce it was 26.5 (±18) and 10.6 (±3.4) mg/L in fe-tive of differences in rate and extent of absorption as well as male and male rats. These plasma concentrations were derivedmetabolism and renal clearance. from the reported blood concentrations in the original study, based The human plasma clearance data used in deriving the blood- on volume adjustment (plasma = 40 ml/kg, blood = 64 ml/kg). Not-based BE for triclosan (0.041 L/kg h, or 2.9 L/kg d) was also ob- ing the loss of body weight towards the end of the 2-year rat study,tained from Sandborgh-Englund et al. (2006). In this study, the EC (2009) used the estimated plasma values associated with the in-subject-speciﬁc values of clearance ranged from 0.032 to 0.049 L/ terim sacriﬁce. Given that the NOAEL in male rats of 12 mg/kg/dh/kg, following a single oral dose of 4 mg/day (which was swal- used in the EU assessment, we chose the corresponding plasma va-lowed completely, contrary to 5–40% fraction swallowed under lue of 21.8 mg/L (intrim sacriﬁce value) as the basis for developingnormal human usage of products containing triclosan – such as the BE, with its associated uncertainty as indicated above.toothpaste) (reviewed in Sandborgh-Englund et al., 2006). Theclearance values reported by Sandborgh-Englund et al. (2006) 4.2. Conﬁdence assessmentand used in the present study correspond to plasma clearance di-vided by the fraction of dose absorbed, and thus they incorporate The guidelines for derivation of BE values (Hays et al., 2008)variability in the extent of absorption in study subjects. specify consideration of two main elements in the assessment of The available data do not support any age or gender related conﬁdence in the derived BE values: robustness of the availablechanges in the pharmacokinetics of triclosan. For example, the pharmacokinetic data and models, and understanding of the rela-plasma half-lives in adults and children estimated in various stud- tionship between the measured biomarker and the critical or rele-ies using toothpaste, dental slurry capsules or aqueous solution vant target tissue dose metric.were comparable, and ranged between 13.4 and 21 h (EC, 2009). For urine-based BE, the cumulative fraction of 54% (at 96 h)The pharmacokinetic study of Sandborgh-Englund et al. (2006) used in the present study, based on data obtained in ten volunteersdid not ﬁnd any consistent gender difference among adults, in receiving a single oral dose of 4 mg (Sandborgh-Englund et al.,either clearance or plasma half-life of triclosan. 2006), is comparable to the observations (44–57% at 48 h) in a Both the plasma- and urine-based BE values for triclosan were number of different clinical and preclinical studies summarizedderived using clearance and excretion fraction data obtained in by DeSalva et al. (1989). However, the relevance of the total triclo-single oral dose studies of Sandborgh-Englund et al. (2006). Oral san level in urine to the target tissue exposure to the toxic form ofroute is the route of exposure in the critical toxicity studies; fur- the chemical is not known. Thus, the assessment of the conﬁdencethermore, it the most important route of human exposure to triclo- level in the derived urinary BE values based on these two factors issan based on product use as well as the extent of absorption as follows:(Bagley and Lin, 2000; Moss et al., 2000; Allmyr et al., 2008; Quec-kenberg et al., 2009). The use of pharmacokinetic data from single Robustness of pharmacokinetic data: MEDIUMdose studies would appear to be relevant since the dose-normal- Relevance of biomarker to relevant dose metrics: LOWized AUC was similar for ingestion of triclosan following single ormultiple exposures (EC, 2009). This is supported by the animal The assessment of the conﬁdence level associated with the pla-studies and repeated human exposure studies with toothpaste ma-based BE values is as follows:and soap use, which indicate that steady-state is reached afterabout 7–10 days (EC, 2009). Accordingly, the uncertainty associ- Robustness of pharmacokinetic data: MEDIUMated with the use of the single oral dose human study of Sand- Relevance of biomarker to relevant dose metrics: MEDIUMborgh-Englund et al. (2006) for deriving BEs for triclosan wouldappear to be low. This reﬂects the relevance of the blood or plasma level of triclo- For deriving BE associated with the EC (2009) assessment, in san to the toxic effects (reviewed in NICNAS, 2009); however itaddition to the human clearance data from Sandborgh-Englund does not differentiate between the various forms of triclosan
16 K. Krishnan et al. / Regulatory Toxicology and Pharmacology 58 (2010) 10–17(free vs. conjugated) and their relevance to the mode of action. The Acknowledgmentspharmacokinetic data, speciﬁcally plasma clearance, obtained fromten volunteers (Sandborgh-Englund et al., 2006) is considered to be Funding for this project was provided under a grant from Healthfairly robust, given the additional support from the literature Canada. The views expressed are those of the authors and do notregarding steady-state and inter-individual variability. necessarily reﬂect the views or policies of Health Canada. This BE dossier has undergone an independent peer-review to assure the4.3. Interpretation of biomonitoring data using BE values methods employed here are consistent with the guidelines for der- ivation (Hays et al., 2008) and communication (LaKind et al., 2008) The BE values presented here represent estimates of the 24-h of Biomonitoring Equivalents and that the best available chemical-average concentrations of triclosan in urine that are consistent speciﬁc data was used in calculating the BEs. We thank the variouswith the existing exposure guidance values resulting from the risk reviewers for their insightful suggestions. Prepared under Healthassessments conducted by various governmental agencies as listed Canada Contract 4500195930.in Table 1. These BE values were derived based on current under-standing of the pharmacokinetic properties of these compoundsin humans. These BE values should be regarded as interim screen-ing values that can be updated or replaced if the exposure guidance Referencesvalues are updated or if the scientiﬁc and regulatory communities Adolfsson-Erici, M., Pettersson, M., Parkkonen, J., Sturve, J., 2002. Triclosan, adevelop additional data on acceptable or tolerable concentrations commonly used bactericide found in human milk and in the aquaticin human biological media. environment in Sweden. Chemosphere 46, 1485–1489. The appropriate uses and limitations of BE values have been dis- Allmyr, M., Harden, F., Toms, L.M., Mueller, J.F., McLachlan, M.S., Adolfsson-Erici, M., Sandborgh-Englund, G., 2008. The inﬂuence of age and gender on triclosancussed previously (Aylward and Hays, 2008; Hays and Aylward, concentrations in Australian human blood serum. Sci. Total Environ. 393 (1),2008; Hays et al., 2008). These BE values can be used as a screening 162–167.tool to evaluation population- or cohort-based biomonitoring data Allmyr, M., Panagiotidis, G., Sparve, E., Diczfalusy, U., Sandborgh-Englund, G., 2009. Human exposure to triclosan via toothpaste does not change CYP3A4 activity orin the context of existing risk assessments. Concentrations in ex- plasma concentrations of thyroid hormones. Basic Clin. Phamacol. 105, 339–344.cess of the BE values, but less than the BEPOD values represent med- Aylward, L.L., Hays, S.M., 2008. Biomonitoring equivalents (BE) dossier for 2, 4-ium priority for risk assessment follow-up, while those in excess of dichlorophenoxyacetic acid (2,4D) (CAS No. 94-75-7). Reg. Toxicol. Pharmacol. 51, S37–S48.the BEPOD indicate high priority for risk assessment follow-up. Bagley, D.M., Lin, Y.J., 2000. Clinical evidence for the lack of triclosan accumulationBased on the results of such comparisons, an evaluation can be from daily use in dentifrices. Am. J. Dent. 13, 148–152.made of the need for additional studies on exposure pathways, po- Barr, D.B., Wilder, L.C., Caudill, S.P., Gonzalez, A.J., Needham, L.L., Pirkle, J.L., 2005. Urinary creatinine concentrations in the US population, implications for urinarytential health effects, other aspects affecting exposure or risk, or biologic monitoring measurements. Environ. Health Perspect. 113, 192–200.other risk management activities. Bhargava, H.N., Leonard, P.A., 1996. Triclosan, applications and safety. Am. J. Infect BE values do not represent diagnostic criteria and cannot be Control 24, 209–218.used to evaluate the likelihood of an adverse health effect in an Calafat, A.M., Ye, X., Wong, L.Y., Reidy, J.A., Needham, L.L., 2008. Urinary concentrations of triclosan in the US population 2003–2004. Environ. Healthindividual or even among a population. Measured values in excess Perspect 116, 303–307.of the identiﬁed BE values may indicate exposures at or above the Ciba-Geigy Corporation, 1986. FAT 80’023: 2-Year Oral Administration to Rats.current exposure guidance values that are the basis of the BE der- Unpublished Report No. MIN 83305, Pharmaceuticals Division, Ciba-Geigy Corporation, NJ, USA (as cited in NICNAS (2009)).ivations. However, as discussed above, measured concentrations Ciba-Geigy Limited, 1976. Pharmacokinetic and Metabolic Studies in Man Followingabove the BE values, which are based on 24-h average urinary con- Oral Administration of a 14C-Labelled Preparation. Unpublished Report No: B 6/centrations, would be expected even if exposures do not exceed 1976. Ciba-Geigy Limited, Basel, Switzerland (as cited in NICNAS (2009)). Ciba-Geigy Limited, 1977. Comparison of Pharmacokinetic and Metabolicthe exposure guidance values due to the transient concentration Parameters of Triclosan and HCP in the Mouse, Rat, Beagle Dog and Baboon,proﬁles in urine expected for these compounds, variations in Part A: Survey of Findings, Part B: Detailed Account of the Study. Unpublishedhydration status, and other factors discussed further above. Thus, Report No: B 1/1977. Ciba-Geigy Limited, Basel, Switzerland (as cited in NICNAS (2009)).interpretation of data for individuals or of tails of the distribution Ciba-Geigy Limited, 1990. Report on Partition Coefﬁcient – by OECD TG 107.in population-monitoring studies is not appropriate. Unpublished Test Report: Anal. Test. No. FC-90/1T. Ciba-Geigy Limited, Basel, In addition, the exposure guidance values for triclosan were de- Switzerland. Crofton, K.M., Paul, K.B., DeVito, M.J., Hedge, J.M., 2007. Short-term in vivo exposurerived with a substantial margin from doses that resulted in no ob- to the water contaminant triclosan: evidence for disruption of thyroxine.served effect in the most sensitive animal toxicity studies. Thus, Environ. Toxicol. Pharmacol. 24, 194–197.these values are not ‘‘bright lines” that distinguish safe from unsafe Dayan, A.D., 2007. Risk assessment of triclosan [Irgasan] in human breast milk. Foodexposure levels. Chronic exposure guidance values are set at expo- Chem. Toxicol. 45, 125–129. DeSalva, S.J., Kong, B.M., Lin, Y.J., 1989. Triclosan: a safety proﬁle. Am. J. Dent. 2,sure levels that are expected to be protective over a lifetime of 185–196.exposure. For short-lived compounds such as triclosan, an exceed- Drake, J.C., Buxtorf, A., 1976. 1 Year Oral Toxicity Study in Baboons with Compoundance of the corresponding BE value in a single urine sample may or FAT 80 023/A. Geigy Pharmaceuticals, Toxicology Department, MRID # 133230. Engelhaupt, E., 2007. More Triclosan Trouble. Environmental Science andmay not reﬂect continuing elevated exposure. As demonstrated in Technology. American Chemical Society, p. 2072 (March 1st, 2010) http://the limited available datasets and based on the kinetics of urinary pubs.acs.org/doi/pdfplus/10.1021/es072500z.elimination, spot urinary concentrations may vary substantially European Commission, 2009. Scientiﬁc Committee on Consumer Products (SCCP) Opinion on Triclosan COLIPA No. P32 (May 21th, 2009) http://ec.europa.eu/both within and across days in an individual. Thus, occasional health/ph_risk/committees/04_sccp/docs/sccp_o_166.pdf.exceedances of the BE value in individuals in cross-sectional stud- Garde, A.H., Hansen, A.M., Kristiansen, J., Knudsen, L.E., 2004. Comparison ofies do not imply that adverse health effects are likely to occur, but uncertainties related to standardization of urine samples with volume and creatinine concentration. Ann. Occup. Hyg. 48, 171–179.can serve as an indicator of relative priority for further risk assess- Hanioka, N., Omae, E., Nishimura, T., Jinno, H., Onodera, S., Yoda, R., et al., 1996.ment follow-up. Further discussion of interpretation and commu- Interaction of 2,4,40 -trichloro-20 -hydroxydiphenyl ether with microsomalnication aspects of the BE values is presented in LaKind et al. cytochrome P450-dependent monooxygenases in rat liver. Chemosphere 33, 265–276.(2008) and at www.biomonitoringequivalents.net. Hays, S.M., Aylward, L.L., 2008. Biomonitoring equivalents (BE) dossier for acrylamide (AA) (CAS No. 79–06-1). Reg. Toxicol. Pharmacol. 51, S57–S67.5. Conﬂict of interest statement Hays, S.M., Aylward, L.L., LaKind, J.S., Bartels, M.J., Barton, H.A., Boogaard, P.J., Brunk, C., DiZio, S., Dourson, M., Goldstein, D.A., Lipscomb, J., Kilpatrick, M.E., Krewski, D., Krishnan, K., Nordberg, M., Okino, M., Tan, Y.M., Viau, C., Yager, J.W., 2008. The authors declare they have no conﬂicts of interest. Guidelines for the derivation of biomonitoring equivalents: report from the
K. Krishnan et al. / Regulatory Toxicology and Pharmacology 58 (2010) 10–17 17 biomonitoring equivalents expert workshop. Regul. Toxicol. Pharmacol. 51, S4– Perucca, J., Bouby, N., Valeix, P., Bankir, L., 2007. Sex difference in urine S15. concentration across differing ages, sodium intake, and level of kidneyHays, S.M., Becker, R.A., Leung, H.W., Aylward, L.L., Pyatt, D.W., 2007. disease. Am. J. Physiol. Regul. Integr. Comp. Physiol. 292, R700–R705. Biomonitoring equivalents: a screening approach for interpreting Queckenberg, C., Meins, J., Wachall, B., Doroshyenko, O., Tomalik-Scharte, D., biomonitoring results from a public health risk perspective. Regul. Toxicol. Bastian, B., Abdel-Tawab, M., Fuhr, U., 2009. Absorption, pharmacokinetics and Pharmacol. 47, 96–109. safety of triclosan after dermal administration. Antimicrob. Agents Chemother.Heath, R.J., Rubin, J.R., Holland, D.R., Zhang, E., Snow, M.E., Rock, C.O., 1999. 54, 570–572. Mechanism of triclosan inhibition of bacterial fatty acid synthesis. J. Biol. Chem. Remer, T., Fonteyn, N., Alexy, U., Berkemeyer, S., 2006. Longitudinal examination of 274, 11110–11114. 24-h urinary iodine excretion in schoolchildren as a sensitive, hydration status-Hovander, L., Malmberg, T., Athanasiadou, M., Athanassiadis, I., Rahm, S., Bergman, independent research tool for studying iodine status. Am. J. Clin. Nutr. 83, 639– A., Wehler, E.K., 2002. Identiﬁcation of hydroxylated PCB metabolites and other 646. phenolic halogenated pollutants in human blood plasma. Arch. Environ. Remer, T., Neubert, A., Maser-Gluth, C., 2002. Anthropometry-based reference Contam. Toxicol. 42, 105–117. values for 24-h urinary creatinine excretion during growth and their use inJacobs, M.N., Nolan, G.T., Hood, S.R., 2005. Lignans, bacteriocides and endocrine and nutritional research. Am. J. Clin. Nutr. 75, 561–569. organochlorine compounds activate the human pregnane X receptor (PXR). Samuel, B.U., Hearn, B., Mack, D., Wender, P., Rothbard, J., Kirisits, M.J., Mui, E., Toxicol. Appl. Pharmacol. 209, 123–133. Wernimont, S., Roberts, C.W., Muench, S.P., Rice, D.W., Prigge, S.T., Law, A.B.,LaKind, J.S., Aylward, L.L., Brunk, C., DiZio, S., Dourson, M., Goldstein, D.A., Kilpatrick, McLeod, R., 2003. Delivery of antimicrobials into parasites. Proc. Natl. Acad. Sci. M.E., Krewski, D., Bartels, M.J., Barton, H.A., Boogaard, P.J., Lipscomb, J., Krishnan, USA 55, 14281–14286. K., Nordberg, M., Okino, M., Tan, Y.M., Viau, C., Yager, J.W., Hays, S.M., 2008. Sandborgh-Englund, G., Adolfsson-Erici, M., Odham, G., Ekstrand, J., 2006. Guidelines for the communication of biomonitoring equivalents: report from Pharmacokinetics of triclosan following oral ingestion in humans. J. Toxicol. the biomonitoring equivalents expert workshop. Regul. Toxicol. Pharmacol. 51, Environ. Health A 69, 1861–1873. S16–S26. Sullivan, A., Wretlind, B., Nord, C.E., 2003. Will triclosan in toothpaste select forLevy, C.W., Roujeinikova, A., Sedelnikova, S., Baker, P.J., Stuitje, A.R., Slabas, A.R., resistant oral streptococci? Clin. Microbiol. Infect. 9, 306–309. Rice, D.W., Rafferty, J.B., 1999. Molecular basis of triclosan activity. Nature 398, Surolia, N., Ramachandra Rao, S.P., Surolia, A., 2001. Paradigm shifts in malaria 383–384. parasite biochemistry and anti-malarial chemotherapy. Bioessays 31, 192–196.Mage, D.T., Allen, R.H., Gondy, G., Smith, W., Barr, D.B., Needham, L.L., 2004. USEPA, 2008a. Reregistration Eligibility Decision for Triclosan, (May 21th, 2009). Estimating pesticide dose from urinary pesticide concentration data by http://www.epa.gov/oppsrrd1/REDs/2340red.pdf. creatinine correction in the Third National Health and Nutrition Examination USEPA (United States Environmental Protection Agency), 2008b. Child-speciﬁc Survey (NHANES-III). J. Expo. Anal. Environ. Epidemiol. 14, 457–465. Exposure Factors Handbook. National Center for Environmental Assessment,McGinnis, D., 2008. Toxicological Proﬁle of Triclosan in the Aquatic Environment – Ofﬁce of Research and Development. EPA/600/R-06/096F. Derivation of a National Water Quality Criterion. Department of Marine and Wang, L.Q., Falany, C.N., James, M.O., 2004. Triclosan as a substrate and inhibitor of Environmental Systems, Florida Institute of Technology. www.dalemcginnis. 30 -phosphoadenosine-50 -phosphosulfatesulfotransferase and UDP- com/docs/triclosan.pdf. glucuronosyl transferase in human liver fractions. Drug Metab. Dispos. 32,McMurry, L.M., Oethinger, M., Levy, S.B., 1998. Overexpression of marA, soxS, or 1162–1169. acrAB producesresistance to triclosan in laboratory and clinical strains of Wolff, M.S., Teitelbaum, S.L., Windham, G., Pinney, S.M., Britton, J.A., Chelimo, C., Escherichia coli. FEMS Microbiol. Lett. 166, 305–309. Godbold, J., Biro, F., Kushi, L.H., Pfeiffer, C.M., Calafta, A.M., 2007. Pilot study ofMoss, T., Howes, D., Williams, F.M., 2000. Percutaneous penetration and dermal urinary biomarkers of phytoestrogens, phthalates, and phenols in girls. Environ. metabolism of triclosan (2,4,40 -trichloro-20 -hydroxydiphenyl ether). Food Health Perspect. 115, 116–117. Chem. Toxicol. 38, 361–370. World Health Organization (WHO), 1999. International Programme on ChemicalNational Industrial Chemicals Notiﬁcation and Assessment Scheme (NICNAS), 2009. Safety: Assessing Human Health Risks of Chemicals: Principles for the Australian Government, Department of Health and Ageing. Priority Existing Assessment of Risk to Human Health from Exposure to Chemicals. Chemical Assessment Report No. 30, Triclosan, January (May 21th, 2009). Environmental Health Criteria 210, World Health Organisation, Geneva. http://www.nicnas.gov.au/Publications/CAR/PEC/PEC30/PEC_30_Full_Report_ Yazdankhah, S.P., Scheie, A.A., Høiby, E.A., Lunestad, B.T., Heir, E., Fotland, T.Ø., PDF.pdf. Naterstad, K., Kruse, H., 2006. Triclosan and antimicrobial resistance in bacteria:Perencevich, E.N., Wong, M.T., Harris, A.D., 2001. National and regional an overview. Microb. Drug Resist. 12, 83–90. assessment of the antibacterial soap market: a step toward determining Zorrilla, L.M., Gibson, E.K., Jeffay, S.C., Crofton, K.M., Setzer, W.R., Cooper, R.L., Stoker, the impact of prevalent antibacterial soaps. Am. J. Infect. Control 29, T.E., 2009. The effects of triclosan on puberty and thyroid hormones in male 281–283. Wistar rats. Toxicol. Sci. 107, 56–64.