A comprehensive review of existing toxicity and human exposure data for the ultraviolet filter ensulizole (2-phenylbenzimidazole-5-sulfonic acid) as currently used in over-the-counter sunscreen formulations was conducted. Authorized maximum ensulizole usage levels in consumer end-use products worldwide range from 3% to 8%, with the maximum usage level limited to 4% in the United States, Canada, and Australia. Postmarketing clinical safety studies of ensulizole have reported only occasional local skin effects, none of which were associated with systemic toxicity. Ensulizole has been investigated in vitro, in animal toxicity studies, and in human studies for its pharmacokinetics, pharmacodynamics, and potential toxicological properties. Experimentally determined values of 4% for oral absorption in rats and of 0.26% for dermal absorption in humans were used for risk calculation purposes. There was no evidence of ensulizole bioaccumulation from rat in vivo studies, consistent with its high water solubility and low octanol/water partition coefficient. Ensulizole is not classifiable as an irritant, although local skin irritation with no systemic effects was noted in a 3-month repeated-dose dermal toxicity study in rabbits. Ensulizole is non-(photo)sensitizing, non-phototoxic, and has demonstrated low toxicity in acute (oral, dermal, and intraperitoneal) and subchronic repeated-dose studies in mammalian species. Subchronic 3-month no-observed-adverse-effect levels (NOAELs) were identified at 100 mg/kg/day (dermal rabbit) and 1000 mg/kg/day (oral rat OECD 408 study), the highest doses tested, respectively. Ensulizole is considered non-genotoxic, based on negative in vitro studies. No in vivo genotoxicity or long-term carcinogenicity studies were identified. Carcinogenicity risk is not expected based on the negative genotoxicity data, empirical evidence from repeated-dose toxicity and developmental toxicity studies, and the absence of effects on the androgen, estrogen, thyroid, immune, developmental, or reproductive systems. Based on the selected rat subchronic NOAEL of 1000 mg/kg/day and conservative assumptions for estimating the systemic exposure dose (SED) from the application of sunscreen products, margins of safety (defined as NOAEL/SED) >100 were obtained for ensulizole. Therefore, the available data show that ensulizole does not pose risks to human health when used in sunscreen products at concentrations up to 4%, the permitted maximum usage level in the United States, Canada, and Australia.

A comprehensive review of existing toxicity and human exposure data for the ultraviolet filter avobenzone (butyl methoxydibenzoylmethane) was conducted to assess its safety as currently used in over-the-counter sunscreen formulations. Avobenzone has a suitable safety profile without any clear markers of toxicity or endpoints of concern. There are sufficient clinical studies and in vitro and in vivo toxicity studies in animal models to assess avobenzone's pharmacokinetics, pharmacodynamics, and potential toxicological properties, supportive of its long history of safe use. No harmonized dermal absorption value was available, but the clinical data indicate low percutaneous absorption of avobenzone in humans (≤0.59% of the applied dose). There were no data to characterize the distribution of avobenzone; however, four tentative metabolites of avobenzone have been identified, and limited excretion in urine was demonstrated in human biomonitoring studies. Avobenzone generally did not cause dermal irritation or sensitization, but indications of photoallergy have been reported in clinical case studies. The acute toxicity profile indicated that avobenzone has minimal toxicity. The no-observed-adverse-effect level (NOAEL) for general toxicity from a rat dietary subchronic toxicity study was 450 mg/kg/day. There was no evidence of avobenzone effects on immune tissues or the estrogen, androgen, or thyroid systems. Although there were no formal 2-year carcinogenicity studies for avobenzone, a 90-day dietary exposure study in rats did not show any increase in hyperplasia of any tissue or evidence of cytotoxicity, and avobenzone has not shown any indication of genotoxicity either in vitro or in vivo. Together, this indicates that key events for modes of action for avobenzone are absent and carcinogenicity in humans is unlikely. Based on the selected rat subchronic NOAEL and conservative assumptions for estimating the systemic exposure dose (SED) from the application of sunscreen products, margins of exposure (defined as the ratio of NOAEL to SED) greater than 100 were obtained for avobenzone. Therefore, the available data show that avobenzone is unlikely to pose a risk to human health when used in sunscreen products at concentrations up to the permitted maximum usage levels in the United States and Canada, which is 3%.

The ability to produce direct DNA damage (genotoxicity), which underlies the carcinogenicity of various chemicals, is typically evaluated in a regulatory-approved battery of in vitro tests with potential in vivo follow-up. Growing concerns for animal welfare and implementation of regulations restricting the use of animal testing necessitate the introduction of New Approach Methodologies (NAMs). The avian egg-based (in ovo) models were developed as metabolically competent NAMs capable of bioactivation, detoxication, and elimination of xenobiotics to potentially replace short-term in vivo genotoxicity assays for chemicals that are genotoxic in vitro. These models utilize avian (chicken or turkey) fetal livers for the evaluation of endpoints indicative of DNA damage produced by either direct or indirect mechanisms, the formation of nuclear DNA adducts and strand breaks. Avian embryos have genetic and morphologic resemblance to mammals and can be used for the evaluation of other endpoints including histopathology and genomic profiling. A concordance analysis of 87 and 59 chemicals assessed in the chicken and turkey models, respectively, revealed a stronger correlation with the results from in vivo genotoxicity assays (76% and 67% sensitivity, 79% and 72% specificity for chicken and turkey, respectively) compared to in vitro assays (58% and 56% sensitivity, 45% and 63% specificity for chicken and turkey, respectively). These results demonstrate that in ovo models detect the genotoxic potential of a broader range of compounds compared to in vitro assays with S9 supplementation. In conclusion, fertilized avian egg fetal liver assays offer a promising alternative to traditional in vivo genotoxicity assays.

The Chicken Egg Genotoxicity Assay (CEGA) is an avian egg-based model that utilizes the livers of developing chicken embryo-fetuses to assess the ability of chemicals to produce direct DNA damage. The main goal of the study was to evaluate target tissue exposure and metabolism in the CEGA to assess its suitability as a biologically relevant new approach methodology (NAM) for detecting the genotoxic potential of chemicals. An imaging study using two-photon excitation microscopy after the administration of a fluorescent dye (acridine orange) verified that chemicals following administration into the air sac of the fertilized chicken egg reach the target organ, liver. A metabolism study using liquid chromatography with high resolution mass spectrometry (LC/MS), conducted after the administration of benzo(a)pyrene (B(a)P) according to the CEGA protocol, confirmed the formation of sufficient amounts of reactive metabolite(s) responsible for the genotoxic effects of a parent compound upon reaching the target tissue. Moreover, an RNA sequencing study revealed that B(a)P in embryo-fetal chicken livers significantly upregulated several genes responsible for the activity of the CYP1A1 enzyme, which is critical for the bioactivation of B(a)P. These findings, along with the previously reported DNA damage (i.e., DNA adducts and single-strand breaks) produced by B(a)P in CEGA, support sufficient target tissue exposure to B(a)P and the ability of avian fetal livers to bioactivate B(a)P to a reactive intermediate. Overall, the findings in the study support the conclusion that the CEGA can be considered a robust potential alternative to the animal testing strategy for assessing the genotoxic potential of chemicals.

4-(2-Hydroxyethyl) morpholine (HEM) is widely used as a building block of macromolecules in the manufacture of pharmaceuticals and dietary supplements and could remain as an impurity in the finished products. An evaluation of HEM was conducted to identify endpoints that could be used to determine the point-of-departure (POD) for use in assessing the potential risk from exposure to HEM. No oral repeated dose toxicological studies of appropriate duration were found for HEM. Therefore, suitable analogue(s) were identified. Although oral repeated dose studies were available for the analogues, the studies were not of sufficient duration for use in the assignment of a POD for risk evaluation. Accordingly, the Threshold of Toxicological Concern (TTC) approach, which proposes that a de minimis value can be derived to qualitatively assess risk, was considered for HEM. To determine the appropriate TTC approach (genotoxic or non-genotoxic), the genotoxicity of HEM and its analogues were evaluated. The weight of the evidence indicated that HEM, and the appropriate analogues, are not genotoxic. Considering the chemical structure of HEM, the non-genotoxic Cramer class III TTC value of 1.5 μg/kg bw/day was determined to be appropriate for use in safety assessment of HEM as an impurity in products intended for human consumption.

For genotoxic carcinogens, covalent binding to DNA is a critical initiating event in tumorigenesis. The present research investigated dose-effect relationships of three genotoxic carcinogens representing different structural classes, 2-acetylaminofluorene (2-AAF), benzo[a]pyrene (B[a]P) and quinoline (QUI), to assess the existence of no-observed-effect-levels (NOELs) for the formation of DNA adducts. Carcinogens were administered into the air sac of fertilized turkey eggs over wide dose ranges in three daily injections on days 22 to 24 of incubation. DNA adducts were measured in the fetal turkey livers by the 32P-nucleotide postlabeling (NPL) assay. B[a]P and QUI produced DNA adducts in a dosage-related manner and exhibited NOELs at 0.65 and 0.35 mg/kg bw/day, respectively. In contrast, 2-AAF formed DNA adducts at all tested dosages down to 0.005 mg/kg bw/day. Benchmark dose (BMD) analysis identified the potencies of 2-AAF and QUI to be similar, while B[a]P was the least potent compound. Overall, findings in fetal turkey livers demonstrated that exposure levels to genotoxic compounds that do not result in DNA adducts can exist but are not evident with all carcinogens of this type. The use of mechanistic dose-effect studies for genotoxic endpoints can provide critical information for prioritization of concerns for risk assessment.

The genotoxic and clastogenic/aneugeneic potentials of four α,β-unsaturated aldehydes, 2-phenyl-2-butenal, nona-2-trans-6-cis-dienal, 2-methyl-2-pentenal, and p-methoxy cinnamaldehyde, which are used as fragrance materials, were assessed using the Chicken Egg Genotoxicity Assay (CEGA) and the Hen's egg micronucleus (HET-MN) assay, respectively. Selection of materials was based on their chemical structures and the results of their previous assessment in the regulatory in vitro and/or in vivo genotoxicity test battery. Three tested materials, 2-phenyl-2-butenal, nona-2-trans-6-cis-dienal, and 2-methyl-2-pentenal, were negative in both, CEGA and HET-MN assays. These findings were congruent with the results of regulatory in vivo genotoxicity assays. In contrast, p-methoxy cinnamaldehyde, which was also negative in the in vivo genotoxicity assays, produced evidence of DNA damage, including DNA strand breaks and DNA adducts in CEGA. However, no increase in the micronucleus formation in blood was reported in the HET-MN study. Such variation in responses between the CEGA and HET-MN assay can be attributed to differences in the dosing protocols. Pretreatment with a glutathione precursor, N-acetyl cysteine, negated positive outcomes produced by p-methoxy cinnamaldehyde in CEGA, indicating that difference in response observed in the chicken egg and rodent models can be attributed to rapid glutathione depletion. Overall, our findings support the conclusion that CEGA and/or HET-MN can be considered as a potential alternative to animal testing as follow-up strategies for assessment of genotoxic potential of fragrance materials with evidence of genotoxicity in vitro.

AbstractThis chapter provides an overview of the diverse mechanisms, which are involved in the induction of neoplasia produced by chemical carcinogens, including occupational carcinogens. The key events involved in the process of carcinogenesis are discussed, and the broad classification of carcinogens based on their primary mechanism of action is delineated. Two major classes of carcinogens are discussed: (1) DNA‐reactive (genotoxic) carcinogens, which react with nuclear DNA directly producing DNA damage in the target tissue and (2) epigenetic (nongenotoxic) carcinogens, which either indirectly interact with DNA or affect other regulatory macromolecules altering cellular processes involved in growth control and/or cell death in the target tissue. Examples of human carcinogens are provided. In addition, methods of human risk assessment are discussed.

Commonly consumed foods and beverages can contain chemicals with reported carcinogenic activity in rodent models. Moreover, exposures to some of these substances have been associated with increased cancer risks in humans. Food-borne carcinogens span a range of chemical classes and can arise from natural or anthropogenic sources, as well as form endogenously. Important considerations include the mechanism(s) of action (MoA), their relevance to human biology, and the level of exposure in diet. The MoAs of carcinogens have been classified as either DNA-reactive (genotoxic), involving covalent reaction with nuclear DNA, or epigenetic, involving molecular and cellular effects other than DNA reactivity. Carcinogens are generally present in food at low levels, resulting in low daily intakes, although there are some exceptions. Carcinogens of the DNA-reactive type produce effects at lower dosages than epigenetic carcinogens. Several food-related DNA-reactive carcinogens, including aflatoxins, aristolochic acid, benzene, benzo[a]pyrene and ethylene oxide, are recognized by the International Agency for Research on Cancer (IARC) as causes of human cancer. Of the epigenetic type, the only carcinogen considered to be associated with increased cancer in humans, although not from low-level food exposure, is dioxin (TCDD). Thus, DNA-reactive carcinogens in food represent a much greater risk than epigenetic carcinogens.

Mintlactone (chemical name 3,6-dimethyl-5,6,7,7a-tetrahydro-1-benzofuran-2(4H)-one, CAS Number 13341-72-5) is a fragrance and flavor ingredient with reported uses in many different cosmetics, personal care, and household products. In order to evaluate the genotoxic potential of mintlactone, in vitro and in vivo genotoxicity tests were conducted. Results from bacterial mutagenicity tests varied across different batches of differing purity with positive results observed in TA98 only. An in vivo comet assay was also considered to be positive in livers of female mice but negative in male mice. In contrast, in vitro and in vivo micronucleus tests, as well as 3D skin comet/micronucleus tests, were negative, indicating no chromosomal or DNA damage. The underlying causes for these contradictory results are not clear. It appears that the purity and/or stability of the test material may be an issue. In the absence of dependable scientific information on the purity and/or storage stability of mintlactone, its safety for use as a fragrance ingredient cannot be substantiated.

DNA damage is an established initiating event in the mutagenicity and carcinogenicity of genotoxic chemicals. Accordingly, assessment of this endpoint is critical for chemicals which are being developed for use in humans. To assess the ability of the Chicken Egg Genotoxicity Assay (CEGA) to detect genotoxic pharmaceuticals, a set of 23 compounds with different pharmacological and reported genotoxic effects was tested for the potential to produce nuclear DNA adducts and strand breaks in the embryo-fetal livers using the 32P-nucleotide postlabeling (NPL) and comet assays, respectively. Due to high toxicity, two aneugens, colchicine and vinblastine, and an autophagy inhibitor, hydroxychloroquine, could not be evaluated. Out of the 20 remaining pharmaceuticals, 10 including estrogen modulators, diethylstilbestrol and tamoxifen, antineoplastics cyclophosphamide, etoposide, and mitomycin C, antifungal griseofulvin, local anesthetics lidocaine and prilocaine, and antihistamines diphenhydramine and doxylamine, yielded clear positive outcomes in at least one of the assays. The antihypertensive vasodilator hydralazine and antineoplastics streptozotocin and teniposide, produced only DNA strand breaks, which were not dose-dependent, and thus, the results with these 3 pharmaceuticals were considered equivocal. No DNA damage was detected for 7 compounds, including the purine antagonist 6-thioguanine, antipyretic analgesics acetaminophen and phenacetin, antibiotic ciprofloxacin, antilipidemic clofibrate, anti-inflammatory ibuprofen, and sedative phenobarbital. However, low solubility of these compounds limited dosages tested in CEGA. Overall, results in CEGA were largely in concordance with the outcomes in other systems in vitro and in vivo, indicating that CEGA provides reliable detection of DNA damaging activity of genotoxic compounds. Further evaluations with a broader set of compounds would support this conclusion.

BlueScreen HC is a mammalian cell-based assay for measuring the genotoxicity and cytotoxicity of chemical compounds and mixtures. The BlueScreen HC assay has been utilized at the Research Institute for Fragrance Materials in a safety assessment program as a screening tool to prioritize fragrance materials for higher-tier testing, as supporting evidence when using a read-across approach, and as evidence to adjust the threshold of toxicological concern. Predictive values for the BlueScreen HC assay were evaluated based on the ability of the assay to predict the outcome of in vitro and in vivo mutagenicity and chromosomal damage genotoxicity assays. A set of 371 fragrance materials was assessed in the BlueScreen HC assay along with existing or newly generated in vitro and in vivo genotoxicity data. Based on a weight-of-evidence approach, the majority of materials in the data set were deemed negative and concluded not to have the potential to be genotoxic, while only a small proportion of materials were determined to show genotoxic effects in these assays. Analysis of the data set showed a combination of high positive agreement but low negative agreement between BlueScreen HC results, in vitro regulatory genotoxicity assays, and higher-tier test results. The BlueScreen HC assay did not generate any false negatives, thereby providing robustness when utilizing it as a high-throughput screening tool to evaluate the large inventory of fragrance materials. From the perspective of protecting public health, it is desirable to have no or minimal false negatives, as a false-negative result may incorrectly indicate the lack of a genotoxicity hazard. However, the assay did have a high percentage of false-positive results, resulting in poor positive predictivity of the in vitro genotoxicity test battery outcome. Overall, the assay generated 100% negative predictivity and 3.9% positive predictivity. In addition to the data set of 371 fragrance materials, 30 natural complex substances were evaluated for BlueScreen HC, Ames, and in vitro micronucleus assay, and a good correlation in all three assays was observed. Overall, while a positive result may have to be further investigated, these findings suggest that the BlueScreen HC assay can be a valuable screening tool to detect the genotoxic potential of fragrance materials and mixtures.

Acrylonitrile, an industrial chemical, is a multisite carcinogen in rats and mice, producing tumors in four tissues with barrier function, that is, brain, forestomach, Zymbal’s gland, and Harderian gland. To assess mechanism(s) of action (MoA) for induction of neoplasia and to evaluate whether the findings in rodents are indicative of human hazard, data on the potential key effects produced by acrylonitrile in the four rodent target tissues of carcinogenicity were evaluated. A notable finding was depletion of glutathione in various organs, including two target tissues, the brain, and forestomach, suggesting that this effect could be a critical initiating event. An additional combination of oxidative DNA damage and cytotoxic effects of acrylonitrile and its metabolites, cyanide, and 2-cyanoethylene oxide, could initiate pro-inflammatory signaling and sustained cell and tissue injury, leading to compensatory cell proliferation and neoplastic development. The in vivo DNA-binding and genotoxicity of acrylonitrile has been studied in several target tissues with no compelling positive results. Thus, while some mutagenic effects were reported in acrylonitrile-exposed rodents, data to determine whether this mutagenicity stems from direct DNA reactivity of acrylonitrile are insufficient. Accordingly, the induction of tumors in rodents is consistent primarily with a non-genotoxic MoA, although a contribution from weak mutagenicity cannot be ruled out. Mechanistic data to support conclusions regarding human hazard from acrylonitrile exposure is weak. Comparison of metabolism of acrylonitrile between rodents and humans provide little support for human hazard. Three of the tissues affected in bioassays (forestomach, Zymbal’s gland, and Harderian gland) are present only in rodents, while the brain is anatomically different between rodents and humans, diminishing relevance of tumor induction in these tissues to human hazard. Extensive epidemiological data has not revealed causation of human cancer by acrylonitrile.

Bioassays using rodent liver foci are widely used for the rapid detection of carcinogenic activity of chemicals. The detection of foci of altered hepatocytes after exposure to a test agent for a limited time, e.g., several weeks, or in stop-experiments including a recovery period was proposed nearly 40 years ago and has frequently been used for the quantification of the dose dependence of carcinogenic effects. The quantification of hepatocellular altered foci instead of fully developed hepatocellular tumors allows both a more rapid and more sensitive detection of hepatocarcinogenic effects. The interpretation of the studies must always take into consideration "spontaneous" or cryptogenic carcinogenesis in the rat liver. In contrast to the test for initiating activity, the selection of the appropriate marker reaction for the detection of the induced foci of altered hepatocytes may be crucial for the test for promoting activity, because the phenotype of preneoplastic foci may be modulated by the treatment of the animals.

This chapter describes studies on the characteristics of foci and biochemical changes as early indicators for the assessment of liver cell carcinogenesis. Studies with a number of agents have corroborated the sequential events in the process of hepatocarcinogenesis. It is well established that in the induction of hepatocellular carcinoma in rodents, several distinct cellular lesions precede the development of malignant tumors. Oval cells often proliferate during early stages of hepatocarcinogenesis with different chemicals, particularly aminoazo dyes and aromatic amines in rats and in mice. Decrease of pyruvate kinase type L has immunohistochemically been demonstrated in adenosinetriphosphatase -deficient foci produced by different carcinogens, indicating that alterations in carbohydrate metabolism may be significant in the process of hepatocarcinogenesis. P. Bannasch first reported that nitrosomorpholine-induced foci displayed extensive storage of glycogen and related the occurrence of hepatoma in human cases of glycogen storage disease to changes in carbonhydrate metabolism during liver carcinogenesis.

4-Methylimidazole (4-MeI) is a byproduct formed during the cooking of foods containing carbohydrates and amino acids, including the production of flavors and coloring substances, e.g., class III and IV caramel colors, used in many food products with extensive human exposure. Two-year rodent bioassays via oral exposure conducted by the National Toxicology Program reported evidence of carcinogenicity only in B6C3F1 mice (increased alveolar/bronchial neoplasms). In 2011, the International Agency for Research on Cancer classified 4-MeI as Group 2B, “possibly carcinogenic to humans”. An expert panel was commissioned to assess the genotoxic potential of 4-MeI and the plausibility of a genotoxic mode of action in the formation of lung tumors in mice when exposed to high doses of 4-MeI. The panel defined and used a weight-of-evidence (WOE) approach that included thorough evaluation of studies assessing the genotoxic potential of 4-MeI. The panelists categorized each study, consisting of study weight, degree of technical performance, study reliability, and contribution to the overall WOE. Based on the reviewed studies’ weighted contribution, the panel unanimously concluded that the WOE supports no clear evidence of in vivo genotoxicity of 4-MeI and no association for a genotoxic mode of action in the formation of mouse lung tumors.

This chapter presents the current means of qualitative detection and quantitative evaluation of chemicals involved in cancer causation, promotion and development. It discusses the major advances in the understanding of the complex carcinogenic process, leading to a classification of carcinogens into genotoxic and epigenetic categories. The chapter also presents a new definition of what types and amounts of carcinogens are human cancer risks. Human carcinogens that are genotoxic, and most of them are, reproducibly display activity in most of the short-term tests in vitro. The traditional approach to carcinogen bioassay involved the administration of test chemicals to rodents. However, carcinogens of the electrophilic DNA-reactive type are distinct from other xenobiotics. A chemical labelled 'carcinogen' based on bioassay results alone is in fact only an experimental carcinogen. Toxicologists and those who apply toxicological research results to public protection must have much more cognition of the quantitative aspects and interactive events in carcinogenesis.

In the deliberations over many years on the question of thresholds for the carcinogenicity of chemicals, the dominant paradigm has been the linear no-threshold (LNT) model, derived from concepts formulated in radiation mutagenicity. Based on the analogy with radiation, the key mechanistic assumption underlying the assessment of the dose-effect of chemical-induced carcinogenicity has been that any dose, no matter how low, can lead to induction of mutations, which will result in some risk of neoplasia. The LNT assumption, however, was never well founded and, its application to chemical carcinogens, does not allow for differences in their disposition or mechanisms of action. These mechanisms include DNA-reactivity and epigenetic effects, resulting from very different properties of carcinogens, leading to different dose effects. This review of the research on dose effects of chemical carcinogens administered by repeat dosing for long duration reveals that only some experiments involving what are now recognized as DNA-reactive carcinogens yielded dose effects for induction of tumors which were consistent with the absence of a threshold (for 6/14 chemicals). None of these studies, however, included low doses documented not to produce genetic or other cellular toxicities that underlie carcinogenicity. Otherwise, most dose-effect experiments, including all with epigenetic agents (7), revealed no-observed-effect-levels for tumors, indicative of subthreshold doses. Based on highly informative experimental data, including relevant mechanistic data, it is concluded that no-effect-levels exist for both carcinogen-induced precursor effects and neoplasia. Accordingly, we conclude that, at non-toxic dosages, thresholds exist for the induction of experimental cancer by all types of carcinogens.