Low dose assessment of the carcinogenicity of furan in male F344/N Nctr rats in a 2-year gavage study.

Furan is a volatile organic chemical that is a contaminant in many common foods. Furan is hepatocarcinogenic in mice and rats; however, the risk to humans from dietary exposure to furan cannot be estimated accurately because the lowest tested dose of furan in a 2-year bioassay in rats gave nearly a 100% incidence of cholangiocarcinoma. To provide bioassay data that can be used in preparing risk assessments, the carcinogenicity of furan was determined in male F344/N Nctr rats administered 0, 0.02, 0.044, 0.092, 0.2, 0.44, 0.92, and 2 mg furan/kg body weight (BW) by gavage 5 days/week for 2 years. Exposure to furan was associated with the development of malignant mesothelioma on membranes surrounding the epididymis and on the testicular tunics, with the increase being significant at 2 mg furan/kg BW. There was also a dose-related increase in the incidence of mononuclear cell leukemia, with the increase in incidence being significant at 0.092, 0.2, 0.92, and 2 mg furan/kg BW. Dose-related non-neoplastic liver lesions included cholangiofibrosis, mixed cell foci, basophilic foci, biliary tract hyperplasia, oval cell hyperplasia, regenerative hyperplasia, and cytoplasmic vacuolization. The most sensitive non-neoplastic lesion was cholangiofibrosis, the frequency of which increased significantly at 0.2 mg furan/kg BW.

Temporal concordance between apical and transcriptional points of departure for chemical risk assessment.

The number of legacy chemicals without toxicity reference values combined with the rate of new chemical development is overwhelming the capacity of the traditional risk assessment paradigm. More efficient approaches are needed to quantitatively estimate chemical risks. In this study, rats were dosed orally with multiple doses of six chemicals for 5 days and 2, 4, and 13 weeks. Target organs were analyzed for traditional histological and organ weight changes and transcriptional changes using microarrays. Histological and organ weight changes in this study and the tumor incidences in the original cancer bioassays were analyzed using benchmark dose (BMD) methods to identify noncancer and cancer points of departure. The dose-response changes in gene expression were also analyzed using BMD methods and the responses grouped based on signaling pathways. A comparison of transcriptional BMD values for the most sensitive pathway with BMD values for the noncancer and cancer apical endpoints showed a high degree of correlation at all time points. When the analysis included data from an earlier study with eight additional chemicals, transcriptional BMD values for the most sensitive pathway were significantly correlated with noncancer (r = 0.827, p = 0.0031) and cancer-related (r = 0.940, p = 0.0002) BMD values at 13 weeks. The average ratio of apical-to-transcriptional BMD values was less than two, suggesting that for the current chemicals, transcriptional perturbation did not occur at significantly lower doses than apical responses. Based on our results, we propose a practical framework for application of transcriptomic data to chemical risk assessment.

In vivo measurement of the shape of the tissue-refractive-index correlation function and its application to detection of colorectal field carcinogenesis.

Polarization-gated spectroscopy is an established method to depth-selectively interrogate the structural properties of biological tissue. We employ this method in vivo in the azoxymethane (AOM)-treated rat model to monitor the morphological changes that occur in the field of a tumor during early carcinogenesis. The results demonstrate a statistically significant change in the shape of the refractive-index correlation function for AOM-treated rats versus saline-treated controls. Since refractive index is linearly proportional to mass density, these refractive-index changes can be directly linked to alterations in the spatial distribution patterns of macromolecular density. Furthermore, we found that alterations in the shape of the refractive-index correlation function shape were an indicator of both present and future risk of tumor development. These results suggest that noninvasive measurement of the shape of the refractive-index correlation function could be a promising marker of early cancer development.

A data-based assessment of alternative strategies for identification of potential human cancer hazards.

The two-year cancer bioassay in rodents remains the primary testing strategy for in-life screening of compounds that might pose a potential cancer hazard. Yet experimental evidence shows that cancer is often secondary to a biological precursor effect, the mode of action is sometimes not relevant to humans, and key events leading to cancer in rodents from nongenotoxic agents usually occur well before tumorigenesis and at the same or lower doses than those producing tumors. The International Life Sciences Institute (ILSI) Health and Environmental Sciences Institute (HESI) hypothesized that the signals of importance for human cancer hazard identification can be detected in shorter-term studies. Using the National Toxicology Program (NTP) database, a retrospective analysis was conducted on sixteen chemicals with liver, lung, or kidney tumors in two-year rodent cancer bioassays, and for which short-term data were also available. For nongenotoxic compounds, results showed that cellular changes indicative of a tumorigenic endpoint can be identified for many, but not all, of the chemicals producing tumors in two-year studies after thirteen weeks utilizing conventional endpoints. Additional endpoints are needed to identify some signals not detected with routine evaluation. This effort defined critical questions that should be explored to improve the predictivity of human carcinogenic risk.

Bayesian derivation of an oral cancer slope factor distribution for 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK).

The tobacco-specific nitrosamine (TSNA) 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) is classified by the International Agency for Research on Cancer as a Group 1 carcinogen. Cancer risk assessment in humans exposed to TSNAs largely relies on potency values estimated from animal studies, but available cancer potency values for NNK derived from such studies are conflicting. In this analysis, oral cancer slope factors (CSFo) for NNK were derived according to U.S. Environmental Protection Agency guidelines. An animal study in which rats were exposed to NNK in drinking water was selected as the key study. The multistage-cancer model was fit to the tumor incidence data to determine a point of departure for low dose linear extrapolation, using a benchmark response of 10%. CSFo distributions were then computed using Bayesian methods and Monte Carlo simulation. The resultant CSFo point estimate (BMR/BMDL(10)) was 19.2 (mg/kg day)(-1) based on lung tumor data and 12.2 (mg/kg day)(-1) based on pancreatic tumors. The 95th percentiles of the CSFo distributions were 27.3 and 19.3 (mg/kg day)(-1) based on lung and pancreatic tumors, respectively. The approach using Bayesian methods better accounts for the uncertainty inherent in the values generated using input assumptions and provides for a more robust probabilistic dose-response assessment.

Graph models for phenotype and genotype association between oral mucosa and submandibular gland tumorigenesis in rat.

BACKGROUND: In recent years, success of statistics in field of genetics has been the identification of genes that affect the process of disease. Experimental models using animals enable early stages of tumor development to be studied. The aim of this study was to apply graph models to assess the association between the observed phenotypic changes in rat oral mucosa and induced tumorigenesis in the submandibular gland (SMG). MATERIALS AND METHODS: We studied changes in oncogenes TP53 and bcl-2, histopathological and immunomarker variables in samples of oral mucosa and SMG of Wistar male rats, 60 days old and 180 g in weight, in which tumorigenesis was induced in their SMG by a 0.5% solution of 9,10-dimethyl-1,2-benzanthracene in acetone. A set of linear structural equations were defined, with each formula indicating the response variables and the direct influences. In graph models, saliva was considered as a latent variable. The association was analyzed using Graphical Gaussian Markov models and odd ratios. RESULTS: About 40% of animals treated with 9, 10-dimethyl-1, 2-benzanthracene showed histological alterations in the epithelial basal strata of their oral mucosa only at 150 days. Statistical models indicated a relationship between gene alteration in gene bcl-2 in the SMG and histological changes observed in the oral mucosa (P = 0.04). CONCLUSION: Graph statistical model with one latent variable allows to conclude that these results associated with other clinical parameters may be useful in detecting early changes in SMG tumorigenesis. Furthermore, the design of randomized sampling of oral mucosa allows to validate these results and establish a reliable methodology for presumptive diagnosis or screening in the future.

Estimation of lag time between onset of and death from an occult tumor.

A new statistical method for estimating the lag time between onset of and death from an occult tumor is proposed for data without cause-of-death information. In this method, the survival function for time to tumor onset, tumor-specific survival function, and competing risks survival function are estimated using the maximum likelihood estimates of the parameters. The proposed method utilizes the estimated survival functions and statistically imputed fatal tumors to estimate the lag time. This approach is developed for rodent tumorigenicity assays that have at least one interim sacrifice and a terminal sacrifice. If the data contain cause-of-death information given by pathologists and it is believed to be reliable, it may be used for estimating the lag time. The proposed method is illustrated using a real data set. The accuracy of the estimation of lag time is evaluated via a Monte Carlo simulation study. This study shows that the estimated lag time is quite reliable.

Bayesian estimation of pharmacokinetic and pharmacodynamic parameters in a mode-of-action-based cancer risk assessment for chloroform.

Chloroform is a carcinogen in rodents and its carcinogenicity is secondary to events associated with cytotoxicity and regenerative cell proliferation. In this study, a physiologically based pharmacokinetic/pharmacodynamic (PBPK/PD) model that links the processes of chloroform metabolism, reparable cell damage, cell death, and regenerative cellular proliferation was developed to support a new cancer dose-response assessment for chloroform. Model parameters were estimated using Markov Chain Monte Carlo (MCMC) analysis in a two-step approach: (1) metabolism parameters for male and female mice and rats were estimated against available closed chamber gas uptake data; and (2) PD parameters for each of the four rodent groups were estimated from hepatic and renal labeling index data following inhalation exposures. Subsequently, the resulting rodent PD parameters together with literature values for human age-dependent physiological and metabolism parameters were used to scale up the rodent model to a human model. The human model was used to predict exposure conditions under which chloroform-mediated cytolethality is expected to occur in liver and kidney of adults and children. Using the human model, inhalation Reference Concentrations (RfCs) and oral Reference Doses (RfDs) were derived using an uncertainty factor of 10. Based on liver and kidney dose metrics, the respective RfCs were 0.9 and 0.09 ppm; and the respective RfDs were 0.4 and 3 mg/kg/day.

Data requirements for risk assessment of furan in food.

Furan is an organic, volatile compound used in various chemical-manufacturing industries. Headspace gas chromatography is the analytical method of choice for obtaining reliable results on its occurrence. The presence of furan in some food items has been known since the late 1970s, but a US Food and Drug Administration (FDA) survey published in 2004 revealed the occurrence of furan in a broad variety of canned and jarred foods, including baby food, that undergo heat treatment. Furan is carcinogenic in rats and mice, showing a dose-dependent increase in hepatocellular adenomas and carcinomas. In rats, a dose-dependent increase of mononuclear leukaemia is evident and a very high incidence of cholangiocarcinomas of the liver, even at the lowest dose tested. There is evidence to indicate that furan-induced carcinogenicity is probably attributable to a genotoxic mechanism. However, chronic toxicity with secondary cell proliferation may indirectly amplify the tumour response. From the available data, there is a relative small difference between possible human exposure and the doses in experimental animals required to produce carcinogenic effects. However, reliable risk assessment requires further data on both toxicity and exposure. The European Food Safety Authority's (EFSA) Scientific Panel on Contaminants in the Food Chain (CONTAM) recommended these studies as part of a reliable risk assessment of furan in food.

Approaches to the risk assessment of genotoxic carcinogens in food: a critical appraisal.

The present paper examines the particular difficulties presented by low levels of food-borne DNA-reactive genotoxic carcinogens, some of which may be difficult to eliminate completely from the diet, and proposes a structured approach for the evaluation of such compounds. While the ALARA approach is widely applicable to all substances in food that are both carcinogenic and genotoxic, it does not take carcinogenic potency into account and, therefore, does not permit prioritisation based on potential risk or concern. In the absence of carcinogenicity dose-response data, an assessment based on comparison with an appropriate threshold of toxicological concern may be possible. When carcinogenicity data from animal bioassays are available, a useful analysis is achieved by the calculation of margins of exposure (MOEs), which can be used to compare animal potency data with human exposure scenarios. Two reference points on the dose-response relationship that can be used for MOE calculation were examined; the T25 value, which is derived from linear extrapolation, and the BMDL10, which is derived from mathematical modelling of the dose-response data. The above approaches were applied to selected food-borne genotoxic carcinogens. The proposed approach is applicable to all substances in food that are DNA-reactive genotoxic carcinogens and enables the formulation of appropriate semi-quantitative advice to risk managers.

Review of studies concerning the tumorigenicity of 2-butoxyethanol in B6C3F1 mice and its relevance for human risk assessment.

The U.S. National Toxicology Program (NTP) has completed 2-yr inhalation exposures in rats and mice with 2-butoxyethanol (BE). This review concerns the most significant findings from those studies and describes recent research into the mechanistic aspects of BE-mediated tumorigenesis in the mouse and the relevance of such effects to humans. Two tumor types were increased in B6C3F1 mice leading to the classification of "some evidence" of carcinogenicity: liver hemangiosarcomas in male mice and forestomach tumors in female mice (primarily benign papillomas). The results of research collected to date indicate that the tumorigenesis noted for BE was produced by indirect mechanisms. In particular, the occurrence of liver hemangiosarcomas in male mice has been linked to oxidative damage subsequent to red blood cell hemolysis and iron deposition in this organ. Oral administration of BE in mice up to 600 mg/kg/d for up to 90 d produces a dose-related increase in iron (Perl's staining) in Kupffer cells and hepatocytes, increased DNA synthesis in endothelial cells, and enhanced oxidative damage. Further, iron alone, and not BE or BAA, is responsible for producing oxidative damage in cultured hepatocytes from rats or mice. Forestomach neoplasms in female mice were most likely a result of prolonged exposure-induced irritation with compensatory hyperplasia and subsequent tumor promotion. This mechanism is supported by studies indicating elevated levels of BE and BAA in the mouse forestomach tissues and stomach contents following multiple routes of exposure, forestomach epithelial cell cytotoxicity and cell proliferation following administration of BE and BAA, and the increased capacity of forestomach tissues from female mice to metabolize BE to the more irritating metabolite, BAA. The current article summarizes the results of a number of in vivo and in vitro studies designed to elucidate the underlying mechanisms of tumorigenesis by BE in the mouse and discusses the relevance of these for human risk.

Cancer risk assessment of toxaphene.

The primary purpose is to do cancer risk assessment of toxaphene by using four steps of risk assessment proposed by the United States National Academy of Sciences/National Research Council (NAS/NRC). Four steps of risk assessment including hazard identification, dose-response relationship, exposure assessment, and risk characterization were used to evaluate cancer risk of toxaphene. Toxaphene was the most heavily used insecticide in many parts of the world before it was banned in 1982. It increased incidence of neoplasms of liver and uterus in mice and increased incidence of neoplasms of endocrine organs, thyroid, pituitary, adrenal, mammary glands, and reproductive systems in rats. From mice's and rats' study, slope factor for toxaphene is 0.8557 (mg/ kg/day)(-1). Lifetime average daily dose (LADD) of toxaphene from ambient air, surface water, soil, and fish were 1.08 x 10(-6), 5.71 x 10(-6), 3.43 x 10(-7), and 7.96 x 10(-5) mg/kg/day, respectively. Cancer risk of toxaphene for average exposure is 7.42 x 10(-5). From this study, toxaphene might have carcinogenic risk among humans.

Use of transgenic mice in carcinogenicity hazard assessment.

Determining the carcinogenic potential of materials to which humans have significant exposure is an important, complex and imperfect exercise. Not only are the methods for such determinations protracted, expensive and utilize large numbers of animals, extrapolation of data from such studies to human risk is imprecise. Toxicologists have long recognized these shortcomings but the 2-year chronic rodent study has remained the gold standard. Recent developments in the field of molecular oncology and development of methods to insert or inactivate specific genes in animals have provided the tools with which to develop the next generation of carcinogenicity assays. With improved understanding of oncogene activation and tumor suppressor gene inactivation a number of animal models have been developed to dramatically reduce latency for chemically induced cancers. This has led to the development of shorter carcinogenicity assays. Also, because the spontaneous tumor frequencies in these animals are low during the in-life portion of the study, and studies are terminated well before the health complications of advanced aging are observed, it has been possible to reduce the group sizes and reduce animal usage. FDA's adoption of ICH S1B in 1997, (ICH, 1997) "Testing for the Carcinogenicity of Pharmaceuticals," opened the door for the use of such transgenic models in regulatory toxicology. This presentation reviews the current state of the science and its application to regulatory issues.

Maximum likelihood inference for multivariate frailty models using an automated Monte Carlo EM algorithm.

We present a maximum likelihood estimation procedure for the multivariate frailty model. The estimation is based on a Monte Carlo EM algorithm. The expectation step is approximated by averaging over random samples drawn from the posterior distribution of the frailties using rejection sampling. The maximization step reduces to a standard partial likelihood maximization. We also propose a simple rule based on the relative change in the parameter estimates to decide on sample size in each iteration and a stopping time for the algorithm. An important new concept is acquiring absolute convergence of the algorithm through sample size determination and an efficient sampling technique. The method is illustrated using a rat carcinogenesis dataset and data on vase lifetimes of cut roses. The estimation results are compared with approximate inference based on penalized partial likelihood using these two examples. Unlike the penalized partial likelihood estimation, the proposed full maximum likelihood estimation method accounts for all the uncertainty while estimating standard errors for the parameters.

Carcinogenicity of the aromatic amines: from structure-activity relationships to mechanisms of action and risk assessment.

Aromatic amines represent one of the most important classes of industrial and environmental chemicals: many of them have been reported to be powerful carcinogens and mutagens, and/or hemotoxicants. Their toxicity has been studied also with quantitative structure-activity relationship (QSAR) methods: these studies are potentially suitable for investigating mechanisms of action and for estimating the toxicity of compounds lacking experimental determinations. In this paper, we first summarized the QSAR models for the rodent carcinogenicity of the aromatic amines. The gradation of potency of the carcinogenic amines depended firstly on their hydrophobicity, and secondly on electronic (reactivity, propensity to be metabolically transformed) and steric properties. On the contrary, the difference between carcinogenic and non-carcinogenic aromatic amines depended mainly on electronic and steric properties. These QSARs can be used directly for estimating the carcinogenicity of aromatic amines. A two-step prediction is possible: (1) estimation of yes/no activity; (2) if the answer from step 1 is yes, then prediction of the degree of potency. The QSARs for rodent carcinogenicity were put in a wider context by comparing them with those for: (a) Salmonella mutagenicity; (b) general toxicity; (c) enzymatic reactions; (d) physical-chemical reactions. This comparative QSAR exercise generated a coherent global picture of the action mechanisms of the aromatic amines. The QSARs for carcinogenicity were similar to those for Salmonella mutagenicity, thus pointing to a similar mechanism of action. On the contrary, the general toxicity QSARs (both in vitro and in vivo systems) were mostly based on hydrophobicity, pointing to an aspecific mechanism of action much simpler than that for carcinogenicity and mutagenicity. The oxidation of the amines (first step in the main metabolic pathway leading to carcinogenic and mutagenic species) had identical QSARs in both enzymatic and physical-chemical systems, thus providing evidence for the link between simple chemical reactions and those in biological systems. The results show that it is possible to generate mechanistically and statistically sound QSAR models for rodent carcinogenicity, and indirectly that the rodent bioassay is a reliable source of good quality data.

Modeling multivariate survival data by a semiparametric random effects proportional odds model.

In this article, the focus is on the analysis of multivariate survival time data with various types of dependence structures. Examples of multivariate survival data include clustered data and repeated measurements from the same subject, such as the interrecurrence times of cancer tumors. A random effect semiparametric proportional odds model is proposed as an alternative to the proportional hazards model. The distribution of the random effects is assumed to be multivariate normal and the random effect is assumed to act additively to the baseline log-odds function. This class of models, which includes the usual shared random effects model, the additive variance components model, and the dynamic random effects model as special cases, is highly flexible and is capable of modeling a wide range of multivariate survival data. A unified estimation procedure is proposed to estimate the regression and dependence parameters simultaneously by means of a marginal-likelihood approach. Unlike the fully parametric case, the regression parameter estimate is not sensitive to the choice of correlation structure of the random effects. The marginal likelihood is approximated by the Monte Carlo method. Simulation studies are carried out to investigate the performance of the proposed method. The proposed method is applied to two well-known data sets, including clustered data and recurrent event times data.

Assessment of carcinogenicity of di(2-ethylhexyl)phthalate in a short-term assay using Xpa-/- and Xpa-/-/p53+/- mice.

The potential of Xpa-/- and Xpa-/-/p53+/- mice for short-term carcinogenicity assays was evaluated with di(2-ethylhexyl)phthalate (DEHP). Groups of 15 male and female Xpa-/- mice, received diets containing 0, 1, 500, 3,000, or 6,000 ppm DEHP, and wild-type (WT) and Xpa-/-p53+/-mice 0 or 6,000 ppm DEHP for 39 weeks. Xpa-/-, Xpa-/-/p53+/-, and WT males, fed 2,500 ppm p-cresidine, served as a positive control. In all models, the survival was not altered by DEHP. Increased incidences of nonneoplastic lesions were recorded in testes and kidneys with no apparent difference between the models. The only liver tumors in all models were adenomas in males with no statistically significantly increased incidence. For p-cresidine. the survival was decreased (p < 0.05) only in transgenic models. Statistically significantly increased incidences of nonneoplastic lesions were recorded in the liver, urinary bladder, and nasal cavity in all models, and in kidneys in transgenic models. The only tumors with statistically significantly increased incidence were liver adenomas in transgenic models (XPA: I vs 7: 'XPA/p53': 0 vs 12; WT: 0 vs 5, p = 0.053) and urinary bladder carcinomas in XPA/p53 model (0 vs 7). The negative carcinogenic response to DEHP and the positive response to p-cresidine support the expected sensitivity to genotoxic carcinogens in these transgenic models.

The in vivo rodent test systems for assessment of carcinogenic potential.

A Drug Information Association (DIA) workshop was held in May 2001 to discuss the outcome of the International Life Sciences Institute-Health and Environmental Sciences Institute (ILSI-HESI) project on alternative models for carcinogenicity assessment such as the P53(+/-) and XPA(+/-) knockout mouse models, the RasH2 and Tg.AC transgenic mouse models, and the neonatal mouse model. The "ICH Guideline S1B on Testing for Carcinogenicity of Pharmaceuticals" advocates that carcinogenicity testing of pharmaceuticals, when needed, might be carried out choosing one 2-year rodent carcinogenicity study (rat) plus one other study that supplements the 2-year study and providing additional information that is not readily available from the 2-year study: either (1) a short- or medium-term in vivo rodent test system or (2) a 2-year carcinogenicity study in a second rodent species (mouse). Another topic of discussion was pros and cons of the short- and medium-term models being evaluated by the ILSI-HESI project, in particular the usefulness of the models in relation to results of genotoxicity assays. Further discussions were advised following publication of the ILSI-HESI database, which is expected before the end of 2001. The use of the short- and medium-term rodent test systems were not considered appropriate for the assessment of carcinogenic potential of biotechnology-derived medicinal products.