Nongenotoxic carcinogens in the regulatory environment

Assessment of the mode of action underlying development of forestomach tumors in rodents following oral exposure to ethyl acrylate and relevance to humans

Chronic repeated gavage dosing of high concentrations of ethyl acrylate (EA) causes forestomach tumors in rats and mice. For two decades, there has been general consensus that these tumors are unique to rodents because of: i) lack of carcinogenicity in other organs, ii) specificity to the forestomach (an organ unique to rodents which humans do not possess), iii) lack of carcinogenicity by other routes of exposure, and iv) obvious site of contact toxicity at carcinogenic doses. In 1986, EA was classified as possibly carcinogenic to humans by the International Agency for Research on Cancer (IARC). However, by applying a MOA analyses and human relevance framework assessment, the weight-of-evidence supports a cytotoxic MOA with the following key events: i) bolus delivery of EA to forestomach lumen and subsequent absorption, ii) cytotoxicity likely due to saturation of enzymatic detoxification, iii) chronic regenerative hyperplasia, and iv) spontaneous mutation due to increased cell replication and cell population. Clonal expansion of initiated cells thus results in late onset tumorigenesis. The key events in this 'wound and healing' MOA provide high confidence in the MOA as assessed by evolved Bradford-Hill Criteria. The weight-of-evidence supported by the proposed MOA, combined with a unique tissue that does not exist in humans, indicates that EA is highly unlikely to pose a human cancer hazard.

A review of the genotoxic, mutagenic, and carcinogenic potentials of several lower acrylates

Lower alkyl acrylate monomers include methyl-, ethyl-, n-butyl-, and 2-ethylhexyl acrylate. These acrylates are used in the manufacture of acrylic polymers and copolymers for plastics, food packaging, adhesives, and cosmetic formulations. Although there is limited potential for human environmental exposure, occupational exposure can occur via inhalation and dermal contact. Recently, new genotoxicity data have been generated, along with in silico and in vitro read-cross analyses, for these acrylates. The availability of high-throughput screening (HTS) data through the ToxCast™/Tox21 databases allows for consideration of computational toxicology and organization of these data according to the ten key characteristics of carcinogens. Therefore, we conducted a comprehensive review to evaluate the mechanistic, toxicokinetic, animal, and human data, including HTS data, for characterizing the potential carcinogenicity, mutagenicity, and genotoxicity of these acrylates. Toxicokinetic data demonstrate that these acrylates are metabolized rapidly by carboxylesterase hydrolysis and conjugation with glutathione. HTS data demonstrated an overall lack of bioactivity in cancer-related pathways. Overall, the genotoxicity and mutagenicity data support a cytotoxic, non-genotoxic mechanism for these acrylates. Cancer bioassay studies conducted by the oral, dermal, and inhalation routes in animal models with these acrylates did not show any increase in tumor incidence, with two exceptions. At high doses, and secondary to chronic site-of-contact irritation and corrosion, rodent forestomach tumors were induced by oral gavage dosing with ethyl acrylate, and skin tumors were observed following chronic dermal dosing with 2-ethylhexyl acrylate in C3H/HeJ inbred mice (a strain with deficiencies in wound healing), but not in the outbred NMRI strain. For both dermal and forestomach cancers, tumorigenesis is secondary to high doses and long-term tissue damage, shown to be reversible. With evidence that these chemicals are not genotoxic, and that they cause forestomach and dermal tumors through chronic irritation and regenerative proliferation mechanisms, these acrylates are unlikely to pose a human cancer hazard.

An Adverse Outcome Pathway (AOP) for forestomach tumors induced by non-genotoxic initiating events

The utility of rodent forestomach tumor data for hazard and risk assessment has been examined for decades because humans do not have a forestomach, and these tumors occur by varying modes of action (MOAs). We have used the MOA for ethyl acrylate (EA) to develop an Adverse Outcome Pathway (AOP) for forestomach tumors caused by non-genotoxic initiating events. These tumors occur secondary to site of contact induced epithelial cytotoxicity and regenerative repair-driven proliferation. For EA, the critical initiating event (IE) is epithelial cytotoxicity, and supporting key events (KEs) at the cellular and tissue level are increased cell proliferation (KE1) resulting in sustained hyperplasia (KE2), with the adverse outcome of forestomach papillomas and carcinomas. For EA, a pre-molecular initiating event (pre-MIE) of sustained glutathione depletion is probable. Supporting data from butylated hydroxyanisole (BHA) are also reviewed. Although there may be some variability in the pre-MIEs and IEs for BHA and EA, they share the same KEs, and evidence for BHA confers support for the AOP. Evolved Bradford Hill considerations of biological plausibility, essentiality, and empirical support were evaluated per OECD guidance. Although an MIE is not specifically described, overall confidence in the AOP is high due to well-developed and accepted evidence streams, and the AOP can be used for regulatory applications including hazard identification and risk assessment for chemicals that act by this AOP.

Marketed Human Pharmaceuticals Reported to be Tumorigenic in Rodents

The 1994 U.S. Physicians' Desk Reference reports the results of rodent carcinogenicity tests on 241 pharmaceutical agents. Positive results were reported for 101 agents, with a class label implying a carcinogenic hazard also attached to 9 estrogenic, 4 androgenic, and 3 progestogenic agents. Of the positive agents tested in both rats and mice, 39 of 72 were positive in both species, 22 were positive in rats only, and 11 were positive in mice only. There were few (19) unequivocally genotoxic agents, and only one of these was not reported to be tumorigenic to rodents. The liver was a common target organ in rats and mice. In rats, the other most common target organs for tumors were (in decreasing order) the thyroid, testis, mammary, adrenal, and pituitary. In mice, the only other common target organ was the lung.

Expression of exposure in negative carcinogenicity studies: dose/body weight, dose/body surface area, or plasma concentrations?

Evaluation of positive findings in a rodent carcinogenicity study and the subsequent extrapolation to humans is based on chemical structure, mutagenicity, pharmacology, hormone changes, chronic toxicity, and the nature of the tumors induced. For negative studies, adequacy of exposure may become an issue. The use of plasma concentrations as a metric for exposure assumes that each species responds in a similar manner to a given concentration; data are now available that demonstrate that this is not generally true for carcinogenicity. The use of the body surface area metric (i.e., mg/m2) is a special case of interspecies allometric scaling (i.e., W0.67). For a chemical to be amenable to such scaling in toxicology, it must satisfy 3 criteria: (a) the concentration-time profile of the putative toxicant at the site of action must be governed by a scalable pharmacokinetic process (e.g., glomerular filtration); (b) the mechanism of action and the susceptibility of each species to a given systemic exposure must be the same and, for example, be independent of lifespan, cellular repair mechanism/rate, and so forth; and (c) the biological response must depend only on size (e.g., not on race, strain, gender, age, or parity). Carcinogens rarely, if ever, meet these criteria. An empirical analysis of carcinogenic potency data in rodents and in humans shows that, in general, exposure is best expressed in terms of mg/kg body weight.

High dose levels are not necessary in rodent studies to detect human carcinogens

Guidelines for the conduct of rodent carcinogenicity studies stipulate that when the test substance is administered via the diet, its concentration need not exceed 5% of the diet. Since it is now apparent that human carcinogens are amongst the most potent of rodent carcinogens, it should be possible to detect accurately potential human carcinogens by using only relatively low dose levels in rodent studies. Our analysis of the potency of human carcinogens in rodent studies leads to the conclusion that, even after applying a safety factor of 10, there is no purpose in using dose levels higher than 500 mg/kg body weight or 1% in the diet.

Carcinogenesis

The rationale for identifying carcinogens and their mechanisms of action is twofold. First, carcinogens must be identified so they can be eliminated from our environment. And second, mechanisms must be identified that are involved in the conversion of a normal cell to a cancer cell. This knowledge will offer more approaches to cancer prevention. The ability to prevent many common fatal cancers caused by environmental agents was recognized by an expert committee of the World Health Organization in 1964. Their report represented a consensus of expert opinion at that time. Since that report, some writers have suggested that as many as 80% to 90% of cancers are preventable. More conservative estimates range from 40% to 50%. Preventing only 40% of cancers (the most conservative estimate) would result in saving thousands of lives every year. Therefore, all members of the health care team must take an active role in cancer prevention. The nurse's role in cancer prevention is one of education of the public. As more information about carcinogens become available, nurses need to share that information so that individuals are better able to make healthy choices for themselves that will lead to a lower incidence of cancer, our ultimate goal.

Biologically based modeling in toxicology research

Biologically based modeling can be described as the process by which the specific mechanistic steps governing tissue disposition and toxic action of chemicals are expressed in quantitative terms by a set of equations leading to prediction of the outcome of specific toxicological experiments by computer simulation. These models are useful in risk assessment because their mechanistic biological basis permits the high-to-low dose, route to route and interspecies extrapolation of the tissue disposition and toxic action of chemicals. By far their greatest utility is not as "finished" risk assessment models, but as research tools that convey a quantitative expression of our hypotheses of tissue disposition and toxic action of a chemical. A structured modeling approach to toxicology problems helps identify the data gaps in the areas of chemical disposition and toxic action, thus prioritizing on-going research to obtain critical information required to conduct quantitative risk assessment. This paper examines progress in developing comprehensive biologically based models for cancer induction by non-genotoxic carcinogens that are cytotoxic in target tissues. The strategies for linking the models on dosimetry, cytotoxicity, and carcinogenicity are described in detail. The basic concepts and approaches discussed here can be applied to many other toxic chemicals and to toxicity endpoints other than cancer.

Relationship between hepatic peroxisome proliferation and 8-hydroxydeoxyguanosine formation in liver DNA of rats following long-term exposure to three peroxisome proliferators; di(2-ethylhexyl) phthalate, aluminium clofibrate and simfibrate

To elucidate the relationship between hepatic peroxisome proliferation and oxidative DNA damage induced by hepatocarcinogenic peroxisome proliferators, 3 agents, namely, di(2-ethylhexyl) phthalate (DEHP, aluminium clofibrate and simfibrate were fed at doses of 1.2%, aluminium clofibrate 0.5% and 0.5% in the diet, respectively, to male F-344 rats for up to 1 year. Evidence of hepatic peroxisome proliferation and 8-hydroxydeoxyguanosine (8-OH-dG) formation in liver and kidney DNA were assessed at 1, 2, 3, 6, 9 and 12 months. Peroxisomal beta-oxidation enzyme activities were increased 3- to 8-fold and catalase was elevated to 1.4- to 2.2-fold the control level by DEHP, aluminium clofibrate and simfibrate from months 1 to 12 of the treatment. 8-OH-dG levels in liver DNA of DEHP-, aluminium clofibrate- and simfibrate-fed rats were increased approximately 2-fold after 1 month, the tendency for elevation also being observed in the liver DNA at 2, 3, 9 and 12 months. The results thus clearly demonstrate that persistent peroxisome proliferation in the liver leads to continued specific oxidative DNA damage.

Naturally-occurring age-dependent glutathione S-transferase pi immunoreactive hepatocytes in aging female F344 rat liver as potential promotable targets for non-genotoxic carcinogens

Naturally occurring basophilic foci (focal hepatocellular proliferative lesions) (FHPL) in the livers of aging female F344/NCr rats could not be promoted to grow or progress into tumors after phenobarbital (PB) exposure. Instead, PB induced new unique eosinophilic hepatocellular foci and adenomas much quicker in old rats than in young rats. We now report that these foci are immunoreactive for glutathione S-transferase pi (GSTP) and that they appear to arise from some naturally occurring single and double GSTP-reactive cells and foci which occur spontaneously in the liver of aging F344 rats in an age-related fashion. PB and other nongenotoxic chemicals may act as "carcinogens" by promoting the growth (clonal expansion) of some of these putative spontaneously-initiated cells and foci into tumors.

Induction of gastrointestinal tumors in mice fed the fungicide folpet: possible mechanisms

Dietary administration of the fungicide folpet, N-(trichloromethylthio) phthalimide, to B6C3F1 mice at dose levels of 1,000, 5,000 and 10,000 ppm induced a dose-related appearance of duodenal atypical hyperplasia, adenomas and adenocarcinomas. The appearance in some of these animals of gastric papillomas and gastric squamous cell carcinomas was correlated in many instances to the presence of duodenal obstructions. It is suggested that the gastric lesions appeared subsequent to, and as an indirect result of, these partial lumenal duodenal obstructions. We suggest that the presence of duodenal obstructions is consistent with the notion that reflux of folpet, bile acids and pancreatic enzymes into the stomach may have acted to irritate and consequently stimulate local neoplastic proliferation. In addition, the duodenal obstructions may have resulted in delayed emptying time of the stomach contents with consequential stagnation. This would cause high concentrations of folpet to act locally on the gastric mucosa.

Risk assessment of carcinogenic and noncarcinogenic chemicals

"Risk Assessment" is a general term used with increasing frequency by both scientists and regulators. Scientifically based risk assessments consider available toxicologic data when judging which agents pose a significant risk to the human population. The science of toxicology focuses on identifying potential hazards to human health using surrogate animal studies. Margins of Safety and establishment of ADIs (Acceptable Daily Intakes) are methods applied to animal test data to set "safe" levels of potential exposure. While the use of Safety Factors in development of the ADI can support a pragmatic conclusion of safety, this approach cannot provide estimates of the probability of harm or the degree of safety. Therefore, Quantitative Risk Assessment (QRA) methods using mathematical models have been advanced to extrapolate from animal exposures which are usually high to much lower human exposure levels where experimental response is absent. Such methodology has been applied primarily by U.S. regulatory agencies to experimental oncogenic responses to estimate the risks of chemical exposure. The present manuscript considers both methods for evaluation of chemical safety and focuses on the scientific merits and limitations of each.

Carcinogenicity of polyhalogenated biphenyls: PCBs and PBBs

Polychlorinated biphenyls (PCBs) and polybrominated biphenyls (PBBs) are compounds whose physical/chemical properties led to their widespread commercial use. Although their production has been banned or severely limited in most countries since the 1970s, the persistence and stability of these compounds have resulted in a worldwide distribution, especially of PCBs. PBB contamination is limited principally to the state of Michigan, where a series of tragic errors eventually resulted in the accumulation of residues in livestock and the general human population. Long-term exposure to PCBs and PBBs in animals has been associated with the induction of neoplastic nodules in the liver and in some cases hepatocellular carcinoma. This review discusses the role of PCBs and PBBs in the process of carcinogenesis. The mutagenicity/genotoxicity of these compounds, as well as their initiation/promotion potential is discussed. The epidemiology of PCB and PBB exposure is reported along with an estimation of the risk of cancer to humans. Finally, possible molecular mechanisms of action are suggested for polyhalogenated biphenyls in cancer development.