Benchmark dose (BMD) modeling: current practice, issues, and challenges.

Benchmark dose (BMD) modeling is now the state of the science for determining the point of departure for risk assessment. Key advantages include the fact that the modeling takes account of all of the data for a particular effect from a particular experiment, increased consistency, and better accounting for statistical uncertainties. Despite these strong advantages, disagreements remain as to several specific aspects of the modeling, including differences in the recommendations of the US Environmental Protection Agency (US EPA) and the European Food Safety Authority (EFSA). Differences exist in the choice of the benchmark response (BMR) for continuous data, the use of unrestricted models, and the mathematical models used; these can lead to differences in the final BMDL. It is important to take confidence in the model into account in choosing the BMDL, rather than simply choosing the lowest value. The field is moving in the direction of model averaging, which will avoid many of the challenges of choosing a single best model when the underlying biology does not suggest one, but additional research would be useful into methods of incorporating biological considerations into the weights used in the averaging. Additional research is also needed regarding the interplay between the BMR and the UF to ensure appropriate use for studies supporting a lower BMR than default values, such as for epidemiology data. Addressing these issues will aid in harmonizing methods and moving the field of risk assessment forward.

Characterizing Risk for Cumulative Risk Assessments.

In assessing environmental health risks, the risk characterization step synthesizes information gathered in evaluating exposures to stressors together with dose-response relationships, characteristics of the exposed population, and external environmental conditions. This article summarizes key steps of a cumulative risk assessment (CRA) followed by a discussion of considerations for characterizing cumulative risks. Cumulative risk characterizations differ considerably from single chemical- or single source-based risk characterization. CRAs typically focus on a specific population instead of a pollutant or pollutant source and should include an evaluation of all relevant sources contributing to the exposures in the population and other factors that influence dose-response relationships. Second, CRAs may include influential environmental and population-specific conditions, involving multiple chemical and nonchemical stressors. Third, a CRA could examine multiple health effects, reflecting joint toxicity and the potential for toxicological interactions. Fourth, the complexities often necessitate simplifying methods, including judgment-based and semi-quantitative indices that collapse disparate data into numerical scores. Fifth, because of the higher dimensionality and potentially large number of interactions, information needed to quantify risk is typically incomplete, necessitating an uncertainty analysis. Three approaches that could be used for characterizing risks in a CRA are presented: the multiroute hazard index, stressor grouping by exposure and toxicity, and indices for screening multiple factors and conditions. Other key roles of the risk characterization in CRAs are also described, mainly the translational aspect of including a characterization summary for lay readers (in addition to the technical analysis), and placing the results in the context of the likely risk-based decisions.

Evaluation of a Proportional Response Addition Approach to Mixture Risk Assessment and Predictive Toxicology Using Data on Four Trihalomethanes from the U.S. EPA's Multiple-Purpose Design Study.

In this study, proportional response addition (Prop-RA), a model for predicting response from chemical mixture exposure, is demonstrated and evaluated by statistically analyzing data on all possible binary combinations of the four regulated trihalomethanes (THMs). These THMs were the subject of a multipurpose toxicology study specifically designed to evaluate Prop-RA. The experimental design used a set of doses common to all components and mixtures, providing hepatotoxicity data on the four single THMs and the binary combinations. In Prop-RA, the contribution of each component to mixture toxicity is proportional to its fraction in the mixture based on its response at the total mixture dose. The primary analysis consisted of 160 evaluations. Statistically significant departures from the Prop-RA prediction were found for seven evaluations, with three predications that were greater than and four that were less than the predicted response; interaction magnitudes (n-fold difference in response vs. prediction) ranged from 1.3 to 1.4 for the former and 2.6 to 3.8 for the latter. These predictions support the idea that Prop-RA works best with chemicals where the effective dose ranges overlap. Prop-RA does not assume the similarity of toxic action or independence, but it can be applied to a mixture of components that affect the same organ/system, with perhaps unknown toxic modes of action.

Evaluation of the Interaction-Based Hazard Index Formula Using Data on Four Trihalomethanes from U.S. EPA's Multiple-Purpose Design Study.

The interaction-based hazard index (HIINT), a mixtures approach to characterizing toxicologic interactions, is demonstrated and evaluated by statistically analyzing data on four regulated trihalomethanes (THMs). These THMs were the subject of a multipurpose toxicology study specifically designed to evaluate the HIINT formula. This HIINT evaluation uses single, binary and quaternary mixture THM data. While this research is considered preliminary, the results provide insights on the application of HIINT when toxicology mixture data are available and on improvements to the method. The results for relative liver weight show the HIINT was generally not conservative but did adjust the additive hazard index (HI) in the correct direction, predicting greater than dose-additivity, as seen in the mixture data. For the liver serum enzyme endpoint alanine aminotransferase, the results were mixed, with some indices giving an estimated effective dose lower than the observed mixture effective dose and others higher; in general, the HIINT adjusted the HI in the correct direction, predicting less than dose-additivity. In addition, a methodological improvement was made in the calculation of maximum interaction magnitude. Suggested refinements to the HIINT included mixture-specific replacements for default parameter values and approaches for supplementing the usual qualitative discussions of uncertainty with numerical descriptions.

Identification of developmentally toxic drinking water disinfection byproducts and evaluation of data relevant to mode of action.

Reactions between chemicals used to disinfect drinking water and compounds present in source waters produce chemical mixtures containing hundreds of disinfection byproducts (DBPs). Although the results have been somewhat inconsistent, some epidemiological studies suggest associations may exist between DBP exposures and adverse developmental outcomes. The potencies of individual DBPs in rodent and rabbit developmental bioassays suggest that no individual DBP can account for the relative risk estimates reported in the positive epidemiologic studies, leading to the hypothesis that these outcomes could result from the toxicity of DBP mixtures. As a first step in a mixtures risk assessment for DBP developmental effects, this paper identifies developmentally toxic DBPs and examines data relevant to the mode of action (MOA) for DBP developmental toxicity. We identified 24 developmentally toxic DBPs and four adverse developmental outcomes associated with human DBP exposures: spontaneous abortion, cardiovascular defects, neural tube defects, and low birth weight infancy. A plausible MOA, involving hormonal disruption of pregnancy, is delineated for spontaneous abortion, which some epidemiologic studies associate with total trihalomethane and bromodichloromethane exposures. The DBP data for the other three outcomes were inadequate to define key MOA steps.

An approach for assessing human exposures to chemical mixtures in the environment.

Humans are exposed daily to multiple chemicals, including incidental exposures to complex chemical mixtures released into the environment and to combinations of chemicals that already co-exist in the environment because of previous releases from various sources. Exposures to chemical mixtures can occur through multiple pathways and across multiple routes. In this paper, we propose an iterative approach for assessing exposures to environmental chemical mixtures; it is similar to single-chemical approaches. Our approach encompasses two elements of the Risk Assessment Paradigm: Problem Formulation and Exposure Assessment. Multiple phases of the assessment occur in each element of the paradigm. During Problem Formulation, analysts identify and characterize the source(s) of the chemical mixture, ensure that dose-response and exposure assessment measures are concordant, and develop a preliminary evaluation of the mixture's fate. During Exposure Assessment, analysts evaluate the fate of the chemicals comprising the mixture using appropriate models and measurement data, characterize the exposure scenario, and estimate human exposure to the mixture. We also describe the utility of grouping the chemicals to be analyzed based on both physical-chemical properties and an understanding of environmental fate. In the article, we also highlight the need for understanding of changes in the mixture composition in the environment due to differential transport, differential degradation, and differential partitioning to other media. The Results section describes the application of the method to various chemical mixtures, highlighting issues associated with assessing exposures to chemical mixtures in the environment.

Evaluating quantitative formulas for dose-response assessment of chemical mixtures.

Risk assessment formulas are often distinguished from dose-response models by being rough but necessary. The evaluation of these rough formulas is described here, using the example of mixture risk assessment. Two conditions make the dose-response part of mixture risk assessment difficult, lack of data on mixture dose-response relationships, and the need to address risk from combinations of chemicals because of public demands and statutory requirements. Consequently, the U.S. Environmental Protection Agency has developed methods for carrying out quantitative dose-response assessment for chemical mixtures that require information only on the toxicity of single chemicals and of chemical pair interactions. These formulas are based on plausible ideas and default parameters but minimal supporting data on whole mixtures. Because of this lack of mixture data, the usual evaluation of accuracy (predicted vs. observed) cannot be performed. Two approaches to the evaluation of such formulas are to consider fundamental biological concepts that support the quantitative formulas (e.g., toxicologic similarity) and to determine how well the proposed method performs under simplifying constraints (e.g., as the toxicologic interactions disappear). These ideas are illustrated using dose addition and two weight-of-evidence formulas for incorporating toxicologic interactions.

Component-based and whole-mixture techniques for addressing the toxicity of drinking-water disinfection by-product mixtures.

Chemical disinfection of water is of direct public health benefit as it results in decreased water-borne illness. The chemicals used to disinfect water react with naturally occurring organic matter, bromide, and iodide in the source water, resulting in the formation of disinfection by-products (DBPs). Despite the identification of several hundred DBPs, more than 50% of the mass of total organic halide formed during chlorination remains unidentified. The toxic contribution of the DBPs that are formed and present but not yet chemically identified, the unidentified fraction, has been largely unexplored. A better understanding of the potential for adverse human health consequences associated with exposure to the DBPs present in drinking water will be gained by integration of knowledge on the toxicity of individual DBPs; simple, defined DBP mixtures; complex, environmentally realistic DBP mixtures with partial chemical characterization; and the unidentified fraction.

Synergy and other ineffective mixture risk definitions.

A substantial effort has been spent over the past few decades to label toxicologic interaction outcomes as synergistic, antagonistic, or additive. Although useful in influencing the emotions of the public and the press, these labels have contributed fairly little to our understanding of joint toxic action. Part of the difficulty is that their underlying toxicological concepts are only defined for two chemical mixtures, while most environmental and occupational exposures are to mixtures of many more chemicals. Furthermore, the mathematical characterizations of synergism and antagonism are inextricably linked to the prevailing definition of 'no interaction,' instead of some intrinsic toxicological property. For example, the US EPA has selected dose addition as the no-interaction definition for mixture risk assessment, so that synergism would represent toxic effects that exceed those predicted from dose addition. For now, labels such as synergism are useful to regulatory agencies, both for qualitative indications of public health risk as well as numerical decision tools for mixture risk characterization. Efforts to quantify interaction designations for use in risk assessment formulas, however, are highly simplified and carry large uncertainties. Several research directions, such as pharmacokinetic measurements and models, and toxicogenomics, should promote significant improvements by providing multi-component data that will allow biologically based mathematical models of joint toxicity to replace these pairwise interaction labels in mixture risk assessment procedures.

EPA project-level research strategies for chemical mixtures: targeted research for meaningful results.

Project-level research strategies at the U.S. Environmental Protection Agency regarding chemical mixtures are impacted by administrative priorities, public interests, expert opinions, scientific advances, regulatory needs, and legislative actions, influencing the setting of priorities and goals. Perhaps, the most significant influence on conducting chemical mixtures research is the passage of laws requiring the EPA to investigate the potential toxicity of various mixtures, specifically the Comprehensive Environmental Response, Compensation, and Liability Act of 1980, the Food Quality Protection Act of 1996, and the Safe Drinking Water Act Amendments of 1996. Scarce resources are allocated to broadly defined issues for consideration by teams of scientists, who design and implement specific projects. Because resources are limited, projects may have several goals, e.g., filling specific data gaps to support a regulation and, simultaneously, producing data to evaluate a risk assessment method. Research areas of emphasis are shaped by risk assessment needs, data gap uncertainties, and experimental design considerations. This paper discusses factors shaping EPA research strategies for chemical mixtures and presents an example of efficient research planning to investigate potential toxicity from exposure to drinking water disinfection by-products.

A four-step approach to evaluate mixtures for consistency with dose addition.

Mixture risk assessment is often hampered by the lack of dose-response information on the mixture being assessed, forcing reliance on component formulas such as dose addition. We present a four-step approach for evaluating chemical mixture data for consistency with dose addition for use in supporting a component based mixture risk assessment. Following the concepts in the U.S. EPA mixture risk guidance (U.S. EPA, 2000a,b), toxicological interaction for a defined mixture (all components known) is departure from a clearly articulated definition of component additivity. For the common approach of dose additivity, the EPA guidance identifies three desirable characteristics, foremost of which is that the component chemicals are toxicologically similar. The other two characteristics are empirical: the mixture components have toxic potencies that are fixed proportions of each other (throughout the dose range of interest), and the mixture dose term in the dose additive prediction formula, which we call the combined prediction model (CPM), can be represented by a linear combination of the component doses. A consequent property of the proportional toxic potencies is that the component chemicals must share a common dose-response model, where only the dose coefficients depend on the chemical components. A further consequence is that the mixture data must be described by the same mathematical function ("mixture model") as the components, but with a distinct coefficient for the total mixture dose. The mixture response is predicted from the component dose-response curves by using the dose additive CPM and the prediction is then compared with the observed mixture results. The four steps are to evaluate: (1) toxic proportionality by determining how well the CPM matches the single chemical models regarding mean and variance; (2) fit of the mixture model to the mixture data; (3) agreement between the mixture data and the CPM prediction; and (4) consistency between the CPM and the mixture model. Because there are four evaluations instead of one, some involving many parameters or dose groups, there are more opportunities to reject statistical hypotheses about dose addition, thus statistical adjustment for multiple comparisons is necessary. These four steps contribute different pieces of information about the consistency of the component and mixture data with the two empirical characteristics of dose additivity. We examine this four-step approach in how it can show empirical support for dose addition as a predictor for an untested mixture in a screening level risk assessment. The decision whether to apply dose addition should be based on all four of those evidentiary pieces as well as toxicological understanding of these chemicals and should include interpretations of the numerical and toxicological issues that arise during the evaluation. This approach is demonstrated with neurotoxicity data on carbamate mixtures.

Cumulative risk assessment toolbox: methods and approaches for the practitioner.

The historical approach to assessing health risks of environmental chemicals has been to evaluate them one at a time. In fact, we are exposed every day to a wide variety of chemicals and are increasingly aware of potential health implications. Although considerable progress has been made in the science underlying risk assessments for real-world exposures, implementation has lagged because many practitioners are unaware of methods and tools available to support these analyses. To address this issue, the US Environmental Protection Agency developed a toolbox of cumulative risk resources for contaminated sites, as part of a resource document that was published in 2007. This paper highlights information for nearly 80 resources from the toolbox and provides selected updates, with practical notes for cumulative risk applications. Resources are organized according to the main elements of the assessment process: (1) planning, scoping, and problem formulation; (2) environmental fate and transport; (3) exposure analysis extending to human factors; (4) toxicity analysis; and (5) risk and uncertainty characterization, including presentation of results. In addition to providing online access, plans for the toolbox include addressing nonchemical stressors and applications beyond contaminated sites and further strengthening resource accessibility to support evolving analyses for cumulative risk and sustainable communities.

Impact of chemical proportions on the acute neurotoxicity of a mixture of seven carbamates in preweanling and adult rats.

Statistical design and environmental relevance are important aspects of studies of chemical mixtures, such as pesticides. We used a dose-additivity model to test experimentally the default assumptions of dose additivity for two mixtures of seven N-methylcarbamates (carbaryl, carbofuran, formetanate, methomyl, methiocarb, oxamyl, and propoxur). The best-fitting models were selected for the single-chemical dose-response data and used to develop a combined prediction model, which was then compared with the experimental mixture data. We evaluated behavioral (motor activity) and cholinesterase (ChE)-inhibitory (brain, red blood cells) outcomes at the time of peak acute effects following oral gavage in adult and preweanling (17 days old) Long-Evans male rats. The mixtures varied only in their mixing ratios. In the relative potency mixture, proportions of each carbamate were set at equitoxic component doses. A California environmental mixture was based on the 2005 sales of each carbamate in California. In adult rats, the relative potency mixture showed dose additivity for red blood cell ChE and motor activity, and brain ChE inhibition showed a modest greater-than additive (synergistic) response, but only at a middle dose. In rat pups, the relative potency mixture was either dose-additive (brain ChE inhibition, motor activity) or slightly less-than additive (red blood cell ChE inhibition). On the other hand, at both ages, the environmental mixture showed greater-than additive responses on all three endpoints, with significant deviations from predicted at most to all doses tested. Thus, we observed different interactive properties for different mixing ratios of these chemicals. These approaches for studying pesticide mixtures can improve evaluations of potential toxicity under varying experimental conditions that may mimic human exposures.

Incorporating nonchemical stressors into cumulative risk assessments.

The role of nonchemical stressors in modulating the human health risk associated with chemical exposures is an area of increasing attention. On 9 March 2011, a workshop titled "Approaches for Incorporating Nonchemical Stressors into Cumulative Risk Assessment" took place during the 50th Anniversary Annual Society of Toxicology Meeting in Washington D.C. Objectives of the workshop included describing the current state of the science from various perspectives (i.e., regulatory, exposure, modeling, and risk assessment) and presenting expert opinions on currently available methods for incorporating nonchemical stressors into cumulative risk assessments. Herein, distinct frameworks for characterizing exposure to, joint effects of, and risk associated with chemical and nonchemical stressors are discussed.

Cholinesterase inhibition and depression of the photic after discharge of flash evoked potentials following acute or repeated exposures to a mixture of carbaryl and propoxur.

Previously, we reported that acute treatment with propoxur or carbaryl decreased the duration of the photic after discharge (PhAD) of flash evoked potentials (FEPs). In the current studies, we compared the effects of acute or repeated exposure to a mixture of carbaryl and propoxur (1:1.45 ratio; propoxur:carbaryl) on the duration of the PhAD and brain ChE activity in Long Evans rats. Animals were exposed (po) either to a single dose (0, 3, 10, 45 or 75 mg/kg), or 14 daily dosages (0, 3, 10, 30, 45 mg/kg), of the mixture. Acute and repeated treatment with 3mg/kg (or greater) of the mixture produced dose-related inhibition of brain ChE activity. Compared to controls, the PhAD duration decreased after acute administration of 75 mg/kg or repeated treatment with 30 mg/kg of the mixture. The linear relationship between the percent of control brain ChE activity and the PhAD duration was similar for both exposure paradigms. Dose-response models for the acute and repeated exposure data did not differ for brain ChE activity or the duration of the PhAD. Repeated treatment with the mixture resulted in slightly less (13-22%) erythrocyte ChE inhibition than acute exposure. Both acute and repeated treatment resulted in dose-additive results for the PhAD duration and less than dose-additive responses (6-16%) for brain ChE activity for the middle range of dosages. Acute treatment resulted in greater than dose-additive erythrocyte ChE inhibition (15-18%) at the highest dosages. In contrast, repeated treatment resulted in less than dose-additive erythrocyte ChE inhibition (16-22%) at the middle dosages. Brain and plasma levels of propoxur and carbaryl did not differ between the acute and repeated dosing paradigms. In summary, a physiological measure of central nervous system function and brain ChE activity had similar responses after acute or repeated treatment with the carbamate mixture, and brain ChE showed only small deviations from dose-additivity. Erythrocyte ChE activity had larger differences between the acute and repeated treatment paradigms, and showed slightly greater deviations from dose-additivity. Because these treatments utilized larger dosages than anticipated environmental exposures, concern for non-additive effects in humans is minimized. The small magnitude of the deviations from dose-additivity also suggest that in the absence of repeated exposure data, results from an acute study of readily reversible carbamate toxicity can be used to estimate the response to repeated daily exposures.