Using quantitative structure-activity relationship modeling to quantitatively predict the developmental toxicity of halogenated azole compounds.

Developmental toxicity is a relevant endpoint for the comprehensive assessment of human health risk from chemical exposure. However, animal developmental toxicity data remain unavailable for many environmental contaminants due to the complexity and cost of these types of analyses. Here we describe an approach that uses quantitative structure-activity relationship modeling as an alternative methodology to fill data gaps in the developmental toxicity profile of certain halogenated compounds. Chemical information was obtained and curated using the OECD Quantitative Structure-Activity Relationship Toolbox, version 3.0. Data from 35 curated compounds were analyzed via linear regression to build the predictive model, which has an R(2) of 0.79 and a Q(2) of 0.77. The applicability domain (AD) was defined by chemical category and structural similarity. Seven halogenated chemicals that fit the AD but are not part of the training set were employed for external validation purposes. Our model predicted lowest observed adverse effect level values with a maximal threefold deviation from the observed experimental values for all chemicals that fit the AD. The good predictability of our model suggests that this method may be applicable to the analysis of qualifying compounds whenever developmental toxicity information is lacking or incomplete for risk assessment considerations.

Dosimetric anchoring of in vivo and in vitro studies for perfluorooctanoate and perfluorooctanesulfonate.

In order to compare between in vivo toxicity studies, dosimetry is needed to translate study-specific dose regimens into dose metrics such as tissue concentration. These tissue concentrations may then be compared with in vitro bioactivity assays to perhaps identify mechanisms relevant to the lowest observed effect level (LOEL) dose group and the onset of the observed in vivo toxicity. Here, we examine the perfluorinated compounds (PFCs) perfluorooctanoate (PFOA) and perfluorooctanesulfonate (PFOS). We analyzed 9 in vivo toxicity studies for PFOA and 13 in vivo toxicity studies for PFOS. Both PFCs caused multiple effects in various test species, strains, and genders. We used a Bayesian pharmacokinetic (PK) modeling framework to incorporate data from 6 PFOA PK studies and 2 PFOS PK studies (conducted in 3 species) to predict dose metrics for the in vivo LOELs and no observed effect levels (NOELs). We estimated PK parameters for 11 combinations of chemical, species, strain, and gender. Despite divergent study designs and species-specific PK, for a given effect, we found that the predicted dose metrics corresponding to the LOELs (and NOELs where available) occur at similar concentrations. In vitro assay results for PFOA and PFOS from EPA's ToxCast project were then examined. We found that most in vitro bioactivity occurs at concentrations lower than the predicted concentrations for the in vivo LOELs and higher than the predicted concentrations for the in vivo NOELs (where available), for a variety of nonimmunological effects. These results indicate that given sufficient PK data, the in vivo LOELs dose regimens, but not necessarily the effects, could have been predicted from in vitro studies for these 2 PFCs.

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.

Developmental toxicity prediction.

Developmental toxicity may be estimated using commercial and noncommercial software that is already available in the market and/or literature, or models may be built from scratch using both commercial and noncommercial software packages. In this chapter, commonly available software programs that can predict the developmental toxicity of chemicals are described. In addition, a method for developing qualitative structure-activity relationship (SAR) models to predict the developmental toxicity of chemicals qualitatively (yes/no prediction) and quantitative structure-activity relationship (QSAR) models to predict quantitative estimates (e.g., LOAEL) of developmental toxicants is also described in this chapter. Additional information described in this chapter include methods to predict physicochemical properties of chemicals that can be used as descriptor variables in the model building process, statistical methods that be used to build QSAR models as well as methods to validate the models that are developed. Most of the methods described in this chapter can be used to develop models for health endpoints other than developmental toxicity as well.

An in silico approach for evaluating a fraction-based, risk assessment method for total petroleum hydrocarbon mixtures.

Both the Massachusetts Department of Environmental Protection (MADEP) and the Total Petroleum Hydrocarbon Criteria Working Group (TPHCWG) developed fraction-based approaches for assessing human health risks posed by total petroleum hydrocarbon (TPH) mixtures in the environment. Both organizations defined TPH fractions based on their expected environmental fate and by analytical chemical methods. They derived toxicity values for selected compounds within each fraction and used these as surrogates to assess hazard or risk of exposure to the whole fractions. Membership in a TPH fraction is generally defined by the number of carbon atoms in a compound and by a compound's equivalent carbon (EC) number index, which can predict its environmental fate. Here, we systematically and objectively re-evaluate the assignment of TPH to specific fractions using comparative molecular field analysis and hierarchical clustering. The approach is transparent and reproducible, reducing inherent reliance on judgment when toxicity information is limited. Our evaluation of membership in these fractions is highly consistent (˜80% on average across various fractions) with the empirical approach of MADEP and TPHCWG. Furthermore, the results support the general methodology of mixture risk assessment to assess both cancer and noncancer risk values after the application of fractionation.

Application of computational toxicological approaches in human health risk assessment. I. A tiered surrogate approach.

Hazard identification and dose-response assessment for chemicals of concern found in various environmental media are typically based on epidemiological and/or animal toxicity data. However, human health risk assessments are often requested for many compounds found at contaminated sites throughout the US that have limited or no available toxicity information from either humans or animals. To address this issue, recent efforts have focused on expanding the use of structure-activity relationships (SAR) approaches to identify appropriate surrogates and/or predict toxicological phenotype(s) and associated adverse effect levels. A tiered surrogate approach (i.e., decision tree) based on three main types of surrogates (structural, metabolic, and toxicity-like) has been developed. To select the final surrogate chemical and its surrogate toxicity value(s), a weight-of-evidence approach based on the proposed decision tree is applied. In addition, a case study with actual toxicity data serves as the evaluation to support our tiered surrogate approach. Future work will include case studies demonstrating the utility of the surrogate approach under different scenarios for data-poor chemicals. In conclusion, our surrogate approach provides a reasonable starting point for identifying potential toxic effects, target organs, and/or modes-of-action, and for selecting surrogate chemicals from which to derive either reference or risk values.

Development of quantitative structure-activity relationship (QSAR) models to predict the carcinogenic potency of chemicals. II. Using oral slope factor as a measure of carcinogenic potency.

The overall risk associated with exposure to a chemical is determined by combining quantitative estimates of exposure to the chemical with their known health effects. For chemicals that cause carcinogenicity, oral slope factors (OSFs) and inhalation unit risks are used to quantitatively estimate the carcinogenic potency or the risk associated with exposure to the chemical by oral or inhalation route, respectively. Frequently, there is a lack of animal or human studies in the literature to determine OSFs. This study aims to circumvent this problem by developing quantitative structure-activity relationship (QSAR) models to predict the OSFs of chemicals. The OSFs of 70 chemicals based on male/female human, rat, and mouse bioassay data were obtained from the United States Environmental Protection Agency's Integrated Risk Information System (IRIS) database. A global QSAR model that considered all 70 chemicals as well as species and/or sex-specific QSARs were developed in this study. Study results indicate that the species and sex-specific QSARs (r(2)>0.8, q(2)>0.7) had a better predictive abilities than the global QSAR developed using data from all species and sexes (r(2)=0.77, q(2)=0.73). The QSARs developed in this study were externally validated, and demonstrated reasonable predictive abilities.

Application of transcriptional benchmark dose values in quantitative cancer and noncancer risk assessment.

The traditional approach for estimating noncancer and cancer reference values in quantitative chemical risk assessment is time and resource intensive. The extent and nature of the studies required under the traditional approach has limited the number of chemicals with published risk assessments. In this study, female mice were exposed for 13 weeks to multiple concentrations of five chemicals that were positive in a 2-year cancer bioassay. Traditional histological and organ weight changes were evaluated, and gene expression microarray analysis was performed on the target tissues. The histological, organ weight changes, and the original tumor incidences in the original cancer bioassay were analyzed using standard benchmark dose (BMD) methods to identify noncancer and cancer points of departure, respectively. The dose-related changes in gene expression were also analyzed using a BMD approach and the responses grouped based on cellular biological processes. A comparison of the transcriptional BMD values with those for the traditional noncancer and cancer apical endpoints showed a high degree of correlation for specific cellular biological processes. For chemicals with human exposure data, the transcriptional BMD values were also used to calculate a margin of exposure. The margins of exposure ranged from 1900 to 54,000. Both the correlation between the BMD values for the transcriptional and apical endpoints and the margin of exposure analysis suggest that transcriptional BMD values may be used as potential points of departure for noncancer and cancer risk assessment.