Toxins in Botanical Drugs and Plant-derived Food and Feed - from Science to Regulation: A Workshop Review.

In September 2022, the 3rd International Workshop on pyrrolizidine alkaloids (PAs) and related phytotoxins was held on-line, entitled 'Toxins in botanical drugs and plant-derived food and feed - from science to regulation'. The workshop focused on new findings about the occurrence, exposure, toxicity, and risk assessment of PAs. In addition, new scientific results related to the risk assessment of alkenylbenzenes, a distinct class of herbal constituents, were presented. The presence of PAs and alkenylbenzenes in plant-derived food, feed, and herbal medicines has raised health concerns with respect to their acute and chronic toxicity but mainly related to the genotoxic and carcinogenic properties of several congeners. The compounds are natural constituents of a variety of plant families and species widely used in medicinal, food, and feed products. Their individual occurrence, levels, and toxic properties, together with the broad range of congeners present in nature, represent a striking challenge to modern toxicology. This review tries to provide an overview of the current knowledge on these compounds and indicates needs and perspectives for future research.

Mechanistic Skin Modeling of Plasma Concentrations of Sunscreen Active Ingredients Following Facial Application.

Sunscreen products constitute two distinct categories. Recreational sunscreens protect against high-intensity, episodic sun exposure, often applied over the entire body. In contrast, facial sunscreen products are designed for sub-erythemal, low-intensity daily sun exposure. Such different exposures necessitate distinctive product safety assessments. Building on earlier methods for predicting dermal disposition, a mechanistic model was developed to simulate plasma concentrations of seven organic sunscreen active ingredients: avobenzone, ensulizole, homosalate, octinoxate, octisalate, octocrylene, and oxybenzone, following facial application. In vitro permeation testing (IVPT) was performed with two different vehicles using a subset of the UV filters. These IVPT results, in addition to previously published IVPT data and published in vivo Maximal Usage Trial (MUsT) data for the UV filters, were used to train the mechanistic dermal model via a Bayesian Markov chain Monte Carlo (MCMC) method. An external validation of the trained model with real-world in vivo datasets demonstrated that the model's predicted UV filter plasma concentrations align well with experimental measurements and capture the observed inter-individual variability. Predictions of steady-state UV filter plasma concentrations under facial application scenarios at 5% concentration and at the maximal allowable concentrations were then generated by the trained model. Oxybenzone had the greatest predicted plasma concentration following facial application. Homosalate and octisalate predictions had high uncertainty associated with the absence of data. Several application scenarios pertaining to avobenzone, ensulizole, octocrylene and octinoxate were identified in which median plasma concentration levels were at 0.5 ng/ml or below when applied in the recreational or facial product. Model limitations include uncertainty in vehicle/water partitioning, formulation metamorphosis, and UV filter systemic clearance, all of which can be refined with additional data. For UV filters, limiting exposure to facial application reduces human safety concerns based on FDA established thresholds.

Enhancement of Skin Permeability Prediction through PBPK Modeling, Bayesian Inference, and Experiment Design.

Physiologically based pharmacokinetic (PBPK) models of skin absorption are a powerful resource for estimating drug delivery and chemical risk of dermatological products. This paper presents a PBPK workflow for the quantification of the mechanistic determinants of skin permeability and the use of these quantities in the prediction of skin absorption in novel contexts. A state-of-the-art mechanistic model of dermal absorption was programmed into an open-source modeling framework. A sensitivity analysis was performed to identify the uncertain compound-specific, individual-specific, and site-specific model parameters that impact permeability. A Bayesian Markov Chain Monte Carlo algorithm was employed to derive distributions of these parameters given in vitro experimental permeability measurements. Extrapolations to novel contexts were generated by simulating the model following its update with samples drawn from the learned distributions as well as parameters that represent the intended scenario. This algorithm was applied multiple times, each using a unique set of permeability measurements sourced under experimental contexts that differ in terms of the compound, vehicle pH, skin sample anatomical site, and the number of compounds under which each subject's skin samples were tested. Among the data sets used in this study, the highest accuracy and precision in the extrapolated permeability was achieved in those that include measurements conducted under multiple vehicle pH levels and in which individual subjects' skin samples are tested under multiple compounds. This work thus identifies factors for consideration in the design of experiments for the purpose of training dermal models to robustly estimate drug delivery and chemical risk.

Use of in vitro metabolism and biokinetics assays to refine predicted in vivo and in vitro internal exposure to the cosmetic ingredient, phenoxyethanol, for use in risk assessment.

A novel approach was developed to help characterize the biokinetics of the cosmetic ingredient, phenoxyethanol, to help assess the safety of the parent and its major stable metabolite. In the first step of this non-animal tiered approach, primary human hepatocytes were used to confirm or refute in silico predicted metabolites, and elucidate the intrinsic clearance of phenoxyethanol. A key result was the identification of the major metabolite, phenoxyacetic acid (PAA), the exposure to which in the kidney was subsequently predicted to far exceed that of phenoxyethanol in blood or other tissues. Therefore, a novel aspect of this approach was to measure in the subsequent step the formation of PAA in the cells dosed with phenoxyethanol that were used to provide points of departure (PoDs) and express the intracellular exposure as the Cmax and AUC24. This enabled the calculation of the intracellular concentrations of parent and metabolite at the PoD in the cells used to derive this value. These concentrations can be compared with in vivo tissue levels to conclude on the safety margin. The lessons from this case study will help to inform the design of other non-animal safety assessments.

Novel Insights into Pyrrolizidine Alkaloid Toxicity and Implications for Risk Assessment: Occurrence, Genotoxicity, Toxicokinetics, Risk Assessment-A Workshop Report.

This paper reports on the major contributions and results of the 2nd International Workshop of Pyrrolizidine Alkaloids held in September 2020 in Kaiserslautern, Germany. Pyrrolizidine alkaloids are among the most relevant plant toxins contaminating food, feed, and medicinal products of plant origin. Hundreds of PA congeners with widespread occurrence are known, and thousands of plants are assumed to contain PAs. Due to certain PAs' pronounced liver toxicity and carcinogenicity, their occurrence in food, feed, and phytomedicines has raised serious human health concerns. This is particularly true for herbal teas, certain food supplements, honey, and certain phytomedicinal drugs. Due to the limited availability of animal data, broader use of in vitro data appears warranted to improve the risk assessment of a large number of relevant, 1,2-unsaturated PAs. This is true, for example, for the derivation of both toxicokinetic and toxicodynamic data. These efforts aim to understand better the modes of action, uptake, metabolism, elimination, toxicity, and genotoxicity of PAs to enable a detailed dose-response analysis and ultimately quantify differing toxic potencies between relevant PAs. Accordingly, risk-limiting measures comprising production, marketing, and regulation of food, feed, and medicinal products are discussed.

A Mechanistic Bayesian Inferential Workflow for Estimation of In Vivo Skin Permeation from In Vitro Measurements.

Computational models can play an integral role in the chemical risk assessment of dermatological products. However, a limitation on the ability of mathematical models to extrapolate from in vitro measurements to in human predictions arises from context-dependence: modeling assumptions made in one setting may not carry over to another scenario. Mechanistic models of dermal absorption relate the skin penetration kinetics of permeants to their partitioning and diffusion across elementary sub-compartments of the skin. This endows them with a flexibility through which specific model components can be adjusted to better reflect dermal absorption in contexts that differ from the in vitro setting, while keeping fixed any context-invariant parameters that remain unchanged in the two scenarios. This paper presents a workflow for predicting in vivo dermal absorption by integrating a mechanistic model of skin penetration with in vitro permeation test (IVPT) measurements. A Bayesian approach is adopted to infer a joint posterior distribution of context-invariant model parameters. By populating the model with samples of context-invariant parameters from this distribution and adjusting context-dependent parameters to suit the in vivo setting, simulations of the model yield estimates of the likely range of in vivo dermal absorption given the IVPT data. This workflow is applied to five compounds previously tested in vivo. In each case, the range of in vivo predictions encompassed the range observed experimentally. These studies demonstrate that the proposed workflow enables the derivation of mechanistically derived upper bounds on dermal absorption for the purposes of chemical risk assessment.

Assessment of Vehicle Volatility and Deposition Layer Thickness in Skin Penetration Models.

Systemic disposition of dermally applied chemicals is often formulation-dependent. Rapid evaporation of the vehicle can result in crystallization of active compounds, limiting their degree of skin penetration. In addition, the choice of vehicle can affect the permeant's degree of penetration into the stratum corneum. The aim of this study is to build a predictive, mechanistic, dermal absorption model that accounts for vehicle-specific effects on the kinetics of permeant transport into skin. An existing skin penetration model is extended to explicitly include the effect of vehicle volatility over time. Using in vitro measurements of skin penetration by chemicals applied in both a saline and an ethanol solvent, the model is optimized to learn two vehicle-specific quantities: the solvent evaporation rate and the extent of permeant deposition into the upper stratum corneum immediately following application. The dermal disposition estimates of the trained model are subsequently compared against those of the original model using further in vitro measurements. The trained model showed a 1.5-fold improvement and a 19-fold improvement in overall goodness of fit among compounds tested in saline and ethanol solvents, respectively. The proposed model structure can thus form a basis for in vitro to in vivo extrapolations of dermal disposition for skin formulations containing volatile components.

Effect of chain length and branching on the in vitro metabolism of a series of parabens in human liver S9, human skin S9, and human plasma.

Parabens are antimicrobial compounds used as preservatives in cosmetics, foods, and pharmaceuticals. Paraben exposure occurs through a variety of routes including dermal absorption, ingestion, and inhalation. Ester bond hydrolysis has been shown to be the predominant biotransformation for this chemical class. Here we evaluated a series of parabens of increasing alkyl chain length and branching in addition to the aryl side chain of phenyl paraben (PhP). We evaluated the parabens under full Michaelis-Menten (MM) parameters to obtain intrinsic clearance values and found different trends between human liver and skin, which correlate with the predominant esterase enzymes in those matrices, respectively. In liver, where carboxylesterase 1 (CES1) is the predominant esterase enzyme, the shorter chain parabens were more readily metabolized, while in skin, where carboxylesterase 2 (CES2) is the predominant esterase enzyme, the longer chain parabens were more readily metabolized. Alkyl chain branching reduced the hydrolysis rates relative to those for the straight chain compounds, while the addition of a phenyl group, as in PhP, showed an increase in hydrolysis, producing the highest observed hydrolysis rate for skin. These data summarize the structure-metabolism relationship for a series of parabens and contribute to the safety assessment of this class of compounds.

Refinement of the Peroxidase Peptide Reactivity Assay and Prediction Model for Assessing Skin Sensitization Potential.

A peptide reactivity assay with an activation component was developed for use in screening chemicals for skin sensitization potential. A horseradish peroxidase-hydrogen peroxide (HRP/P) oxidation system was incorporated into the assay for characterizing reactivity of hapten and pre-/prohapten sensitizers. The assay, named the Peroxidase Peptide Reactivity Assay (PPRA) had a predictive accuracy of 83% (relative to the local lymph node assay) with the original protocol and prediction model. However, apparent false positives attributed to cysteine depletion at relatively high chemical concentrations and, for some chemicals expected to react with the -NH2 group of lysine, little to no depletion of the lysine peptide were observed. To improve the PPRA, cysteine peptide reactions with and without HRP/P were modified by increasing the number of test concentrations and refining their range. In addition, removal of DL-dithiothreitol from the reaction without HRP/P increased cysteine depletion and improved detection of reactive aldehydes and thiazolines without compromising the assay's ability to detect prohaptens. Modification of the lysine reaction mixture by changing the buffer from 0.1 M ammonium acetate buffer (pH 10.2) to 0.1 M phosphate buffer (pH 7.4) and increasing the level of organic solvent from 1% to 25% resulted in increased lysine depletion for known lysine reactive chemicals. Refinement of the prediction model improved the sensitivity, specificity, and accuracy for hazard identification. These changes resulted in significant improvement of the PPRA making it is a reliable method for predicting the skin sensitization potential of all chemicals, including pre-/prohaptens and directly reactive haptens.

Intrinsic relative potency of a series of pyrrolizidine alkaloids characterized by rate and extent of metabolism.

1,2-Unsaturated pyrrolizidine alkaloids (PAs) are sometimes present in foods or herbal supplements/medicines as impurities and pose potential concerns for liver genotoxicity/carcinogenicity. PAs display a strong structure toxicity relationship, however, current regulatory approaches to risk assessment take the precautionary approach of assuming all PAs display the same potency as the most toxic congeners lasiocarpine (LAS) and riddelliine (RID). Here we explore the relative potencies of a series of structurally diverse PAs by measuring DNA adduct formation in vitro in a rat sandwich culture hepatocyte (SCH) cell system. The adducts generated are consistent with those identified in vivo as biomarkers of PA exposure and potential liver-tumor formation. DNA reactive PAs require metabolic activation to form intermediates that bind DNA, therefore, adduct formation is a direct reflection of reactive metabolite formation. Since the area under the concentration versus time curve (AUC) for the depletion of parent PA from the extracellular media is a measure of PA exposure, the ratio of adducts/AUC provides a measure of hepatocyte exposure to DNA-binding metabolites corresponding to an intrinsic potency for DNA adduct formation. Intrinsic potencies relative to potencies for LAS compare well with existing relative potency data further affirming that PA toxicity varies considerably with chemical structure.

Development of growth equations from longitudinal studies of body weight and height in the full term and preterm neonate: From birth to four years postnatal age.

Physiologically based pharmacokinetic (PBPK) models are developed from compound-independent information to describe important anatomical and physiological characteristics of an individual or population of interest. Modeling pediatric populations is challenging because of the rapid changes that occur during growth, particularly in the first few weeks and months after birth. Neonates who are born premature pose several unique challenges in PBPK model development. To provide appropriate descriptions for body weight (BW) and height (Ht) for age and appropriate incremental gains in PBPK models of the developing preterm and full term neonate, anthropometric measurements collected longitudinally from 1,063 preterm and 158 full term neonates were combined with 2,872 cross-sectional measurements obtained from the NHANES 2007-2010 survey. Age-specific polynomial growth equations for BW and Ht were created for male and female neonates with corresponding gestational birth ages of 25, 28, 31, 34, and 40 weeks. Model-predicted weights at birth were within 20% of published fetal/neonatal reference standards. In comparison to full term neonates, postnatal gains in BW and Ht were slower in preterm subgroups, particularly in those born at earlier gestational ages. Catch up growth for BW in neonates born at 25, 28, 31, and 34 weeks gestational age was complete by 13, 8, 6, and 2 months of life (males) and by 10, 6, 5, and 2 months of life (females), respectively. The polynomial growth equations reported in this paper represent extrauterine growth in full term and preterm neonates and differ from the intrauterine growth standards that were developed for the healthy unborn fetus.

In vitro to in vivo extrapolation for high throughput prioritization and decision making.

In vitro chemical safety testing methods offer the potential for efficient and economical tools to provide relevant assessments of human health risk. To realize this potential, methods are needed to relate in vitro effects to in vivo responses, i.e., in vitro to in vivo extrapolation (IVIVE). Currently available IVIVE approaches need to be refined before they can be utilized for regulatory decision-making. To explore the capabilities and limitations of IVIVE within this context, the U.S. Environmental Protection Agency Office of Research and Development and the National Toxicology Program Interagency Center for the Evaluation of Alternative Toxicological Methods co-organized a workshop and webinar series. Here, we integrate content from the webinars and workshop to discuss activities and resources that would promote inclusion of IVIVE in regulatory decision-making. We discuss properties of models that successfully generate predictions of in vivo doses from effective in vitro concentration, including the experimental systems that provide input parameters for these models, areas of success, and areas for improvement to reduce model uncertainty. Finally, we provide case studies on the uses of IVIVE in safety assessments, which highlight the respective differences, information requirements, and outcomes across various approaches when applied for decision-making.

Design, fabrication, and testing of convex reflective diffraction gratings.

The convex reflective diffraction grating is an essential optical component that lends itself to various applications. In this work, we first outline the design principles of convex diffraction gratings from wavefront quality and efficiency perspectives. We then describe a unique fabrication method that allows for the machining of convex diffraction gratings with variable groove structure, which is extendable to rotationally non-symmetric convex diffraction grating substrates. Finally, we demonstrate two quantitative wavefront measurement methods and respective experimental validation.

Development of a physiologically-based pharmacokinetic model of 2-phenoxyethanol and its metabolite phenoxyacetic acid in rats and humans to address toxicokinetic uncertainty in risk assessment.

2-Phenoxyethanol (PhE) has been shown to induce hepatotoxicity, renal toxicity, and hemolysis at dosages ≥ 400 mg/kg/day in subchronic and chronic studies in multiple species. To reduce uncertainty associated with interspecies extrapolations and to evaluate the margin of exposure (MOE) for use of PhE in cosmetics and baby products, a physiologically-based pharmacokinetic (PBPK) model of PhE and its metabolite 2-phenoxyacetic acid (PhAA) was developed. The PBPK model incorporated key kinetic processes describing the absorption, distribution, metabolism and excretion of PhE and PhAA following oral and dermal exposures. Simulations of repeat dose rat studies facilitated the selection of systemic AUC as the appropriate dose metric for evaluating internal exposures to PhE and PhAA in rats and humans. Use of the PBPK model resulted in refinement of the total default UF for extrapolation of the animal data to humans from 100 to 25. Based on very conservative assumptions for product composition and aggregate product use, model-predicted exposures to PhE and PhAA resulting from adult and infant exposures to cosmetic products are significantly below the internal dose of PhE observed at the NOAEL dose in rats. Calculated MOEs for all exposure scenarios were above the PBPK-refined UF of 25.

Extrapolation of systemic bioavailability assessing skin absorption and epidermal and hepatic metabolism of aromatic amine hair dyes in vitro.

Approaches to assess the role of absorption, metabolism and excretion of cosmetic ingredients that are based on the integration of different in vitro data are important for their safety assessment, specifically as it offers an opportunity to refine that safety assessment. In order to estimate systemic exposure (AUC) to aromatic amine hair dyes following typical product application conditions, skin penetration and epidermal and systemic metabolic conversion of the parent compound was assessed in human skin explants and human keratinocyte (HaCaT) and hepatocyte cultures. To estimate the amount of the aromatic amine that can reach the general circulation unchanged after passage through the skin the following toxicokinetically relevant parameters were applied: a) Michaelis-Menten kinetics to quantify the epidermal metabolism; b) the estimated keratinocyte cell abundance in the viable epidermis; c) the skin penetration rate; d) the calculated Mean Residence Time in the viable epidermis; e) the viable epidermis thickness and f) the skin permeability coefficient. In a next step, in vitro hepatocyte Km and Vmax values and whole liver mass and cell abundance were used to calculate the scaled intrinsic clearance, which was combined with liver blood flow and fraction of compound unbound in the blood to give hepatic clearance. The systemic exposure in the general circulation (AUC) was extrapolated using internal dose and hepatic clearance, and Cmax was extrapolated (conservative overestimation) using internal dose and volume of distribution, indicating that appropriate toxicokinetic information can be generated based solely on in vitro data. For the hair dye, p-phenylenediamine, these data were found to be in the same order of magnitude as those published for human volunteers.

Estimation of in vivo dose of dermally applied chemicals leading to estrogen/androgen receptor-mediated toxicity from in vitro data--Illustration with four reproductive toxicants.

We present a quantitative in vitro-in vivo extrapolation framework enabling the estimation of the external dermal exposure dose from in vitro experimental data relevant to a toxicity pathway of interest. The framework adapts elements of the biological pathway altering dose (BPAD) method [Judson et al. Chem Res Toxicol 2011;24:451] to the case of dermal exposure. Dermal doses of four toxicants equivalent to concentrations characterizing their effect on estrogen receptor α or androgen receptor activity in chemical-activated luciferase expression (CALUX) assays are estimated. The analysis shows that dermal BPADs, calculated from one in vitro concentration, can differ by up to a factor of 55, due to the impact applied dose and dermal exposure scenarios can have on skin permeation kinetics. These features should therefore be taken into account in risk assessment of dermally applied chemicals.

Introduction of a methoxymethyl side chain into p-phenylenediamine attenuates its sensitizing potency and reduces the risk of allergy induction.

The strong sensitizing potencies of the most important primary intermediates of oxidative hair dyes, p-phenylenediamine (PPD) and p-toluylenediamine (PTD, i.e. 2-methyl-PPD) are well established. They are considered as the key sensitizers in hair dye allergic contact dermatitis. While modification of their molecular structure is expected to alter their sensitizing properties, it may also impair their color performance. With introduction of a methoxymethyl side chain we found the primary intermediate 2-methoxymethyl-p-phenylenediamine (ME-PPD) with excellent hair coloring performance but significantly reduced sensitizing properties compared to PPD and PTD: In vitro, ME-PPD showed an attenuated innate immune response when analyzed for its protein reactivity and dendritic cell activation potential. In vivo, the effective concentration of ME-PPD necessary to induce an immune response 3-fold above vehicle control (EC3 value) in the local lymph node assay (LLNA) was 4.3%, indicating a moderate skin sensitizing potency compared to values of 0.1 and 0.17% for PPD and PTD, respectively. Finally, assessing the skin sensitizing potency of ME-PPD under consumer hair dye usage conditions through a quantitative risk assessment (QRA) indicated an allergy induction risk negligible compared to PPD or PTD.

A framework incorporating the impact of exposure scenarios and application conditions on risk assessment of chemicals applied to skin.

PURPOSE: 1. To develop a framework for exposure calculation via the dermal route to meet the needs of 21st century toxicity testing and refine current approaches; 2. To demonstrate the impact of exposure scenario and application conditions on the plasma concentration following dermal exposure. METHOD: A workflow connecting a dynamic skin penetration model with a generic whole-body physiologically-based pharmacokinetic (PBPK) model was developed. The impact of modifying exposure scenarios and application conditions on the simulated steady-state plasma concentration and exposure conversion factor was investigated for 9 chemicals tested previously in dermal animal studies which did not consider kinetics in their experimental designs. RESULTS: By simulating the animal study scenarios and exposure conditions, we showed that 7 studies were conducted with finite dose exposures, 1 with both finite and infinite dose exposures (in these 8 studies, an increase in the animal dose resulted in an increase in the simulated steady-state plasma concentrations (C p,ss)), while 1 study was conducted with infinite dose exposures only (an increase in the animal dose resulted in identical C p,ss). Steady-state plasma concentrations were up to 30-fold higher following an infinite dose scenario vs. a finite dose scenario, and up to 40-fold higher with occlusion vs. without. Depending on the chemical, the presence of water as a vehicle increased or decreased the steady-state plasma concentration, the largest difference being a factor of 16. CONCLUSIONS: The workflow linking Kasting's model of skin penetration and whole-body PBPK enables estimation of plasma concentrations for various applied doses, exposure scenarios and application conditions. Consequently, it provides a quantitative, mechanistic tool to refine dermal exposure calculations methodology for further use in risk assessment.

The incorporation of lysine into the peroxidase peptide reactivity assay for skin sensitization assessments.

To establish further a practical quantitative in chemico reactivity assay for screening contact allergens, lysine peptide was incorporated into a liquid chromatography and tandem mass spectrometry-based assay for reactivity assessments of hapten and pre-/pro-hapten chemical sensitizers. Loss of peptide was determined following 24 h coincubation with test chemical using a concentration-response study design. A total of 70 chemicals were tested in discrete reactions with cysteine or lysine peptide, in the presence and absence of horseradish peroxidase-hydrogen peroxide oxidation system. An empirically derived prediction model for discriminating sensitizers from nonsensitizers resulted in an accuracy of 83% for 26 haptens, 19 pre-/pro-haptens, and 25 nonsensitizers. Four sensitizers were shown to selectively react with lysine including two strong/extreme and two weak sensitizers. In addition, seven sensitizers were identified as having higher reactivity toward lysine compared with cysteine. The majority of sensitizing chemicals (27/45) were reactive toward both cysteine and lysine peptides. An estimate of the relative reactivity potency was determined based on the concentration of test chemical that depletes peptide at or above a threshold positive value. Here, we report the use of EC15 as one example to illustrate the use of the model for screening the skin sensitization potential of novel chemicals. Results from this initial assessment highlight the utility of lysine for assessing a chemical's potential to elicit sensitization reactions or induce hypersensitivity. This approach has the potential to qualitatively and quantitatively evaluate an important mechanism in contact allergy for hazard and quantitative risk assessments without animal testing.

1,3-Butadiene: Biomarkers and application to risk assessment.

1,3-Butadiene (BD) is a known rodent and human carcinogen that is metabolized mainly by P450 2E1 to three epoxides, 1,2-epoxy-3-butene (EB), 1,2:3,4-diepoxybutane (DEB) and 1,2-epoxy-3,4-butanediol (EB-diol). The individual epoxides vary up to 200-fold in their mutagenic potency, with DEB being the most mutagenic metabolite. It is important to understand the internal formation of the individual epoxides to assign the relative risk for each metabolite and to understand the molecular mechanisms responsible for major species differences in carcinogenicity. We have conducted extensive exposure-biomarker studies on mice, rats and humans. Using low exposures that range from current occupational levels to human exposures from tobacco smoke has provided evidence that mice are very different from humans, with mice forming ∼200 times more DEB than humans at exposures of 0.1-1.5ppm BD. While no gender differences have been noted in mice and rats for globin adducts or N-7 guanine adducts, female rats and mice had 2-3-fold higher Hprt mutations and DNA-DNA cross-links, suggesting a gender difference in DNA repair. Numerous molecular epidemiology studies have evaluated globin adducts and Hprt mutations, SCEs and chromosomal abnormalities. None of the blinded studies have shown evidence of human genotoxicity at current occupational exposures and studies of globin adducts have shown similar or lower formation of adducts in females than males. If one calculates the EB dose-equivalents for the three species, mice clearly differ from rats and humans, being ∼44 and 174 times greater than rats and humans, respectively. These data provide a scientific basis for improved risk assessment of BD.