Uptake Kinetics,Species Differences,and the Determination of Equivalent Combinations of Air Concentration and Exposure Duration for Assessment of Acute Inhalation Toxicity

For acute inhalation toxicity assessment, I develop a conceptual framework for expressing combinations of intensity (air concentration) and duration that produce equivalent toxicity by examining how the shape of the body-burden uptake curve during a bout of inhalation interacts with various pharmacodynamic measures of the critical body burden needed to produce toxicity. If toxicity depends on attaining a critical tissue concentration, three existing empirical approaches—Haber's Law, the ten Berge equation, and pure air-concentration-dependence—are but local approximations to different parts of an overarching mathematical relationship. The compound-specific half-life of elimination determines the range of durations for which each applies: durations of one half-life or shorter follow Haber's Law, exposures of 4 or more half-lives follow pure air-concentration-dependence, and intermediate durations can be approximated by the ten Berge equation. Better animal-to-human extrapolation is achieved if exposure durations are expressed as number of species-specific half-lives. I consider several alternative pharmacodynamic criteria, such as the dependence of toxicity on time spent above a critical tissue concentration, or on the area under the tissue concentration curve, on the tissue concentration of a toxic metabolite, or on the imbalance of damage and repair processes.     

Estimation of the benchmark dose by structural equation models

While epidemiological data typically contain a multivariate response and often also multiple exposure parameters, current methods for safe dose calculations, including the widely used benchmark approach, rely on standard regression techniques. In practice, dose-response modeling and calculation of the exposure limit are often based on the seemingly most sensitive outcome. However, this procedure ignores other available data, is inefficient, and fails to account for multiple testing. Instead, risk assessment could be based on structural equation models, which can accommodate both a multivariate exposure and a multivariate response function. Furthermore, such models will allow for measurement error in the observed variables, which is a requirement for unbiased estimation of the benchmark dose. This methodology is illustrated with the data on neurobehavioral effects in children prenatally exposed to methylmercury, where results based on standard regression models cause an underestimation of the true risk.     

Use of Toxicological Data in Estimating Reference Values for Risk Assessment

A number of programs within the U.S. Environmental Protection Agency (USEPA) currently set less-than-lifetime exposure limits in addition to the chronic reference dose (RfD) and reference concentration (RfC). A review of procedures within the USEPA for setting reference values suggests that less-thanlifetime reference values should be more routinely developed and captured in the USEPA's online IRIS database where chronic RfDs and RfCs, as well as cancer slope factors, are currently available. A review of standard testing study protocols was conducted to determine what data were available for setting acute, short-term, and longer-term reference values, as well as chronic values. This review was done from the point of view of endpoints assessed for specific organ systems (both structural and functional), life stages covered by exposure and outcome, durations of exposure covered and the outcomes evaluated for each, and evaluation of latency to response and/or reversibility of effects. This review revealed a number of data gaps and research needs, including the need for an acute and/or short-term testing protocol that can be used to set acute and shortterm reference values, a strategy for when to conduct more extensive testing based on initial screening data or other information (e.g., chemical class, pharmacokinetics, mode of action), additonal standard testing guidlines protocols to allow more complete assessment of certain organ systems and life stages, development of pharmacokinetic data for different life stages, toxicity related to aging, and latency to response, particularly long-term latency as a result of developmental exposures. The implications of this review are discussed relative to characterizing hazard data for setting reference values, and the potential effects on uncertainty factors and low-dose extrapolation.     

Extrapolation for Exposure Duration in Oral Toxicity: A Quantitative Analysis of Historical Toxicity Data

For human risk assessment, experimental data often have to be extrapolated for exposure duration, which is generally done by means of default values. The purpose of the present study was twofold. First, to derive a statistical distribution for differences in exposure duration that can be used in a probabilistic concept for combining assessment factors in risk characterization. Second, to obtain insight in the magnitude of the change in No-Observed-Adverse-Effect-Level (NOAEL) with exposure duration, which will lead to more science-based assessment factors for exposure duration. A large historical database, including 198 substances, was consulted. Ratios were calculated for pairs of NOAELs for systemic toxicity from oral toxicity studies with the same species (rats or mice, various strains) and different exposure duration categories. The Geometric Mean (GM), Geometric Standard Deviation (GSD), and the 90th and 95th percentile values were determined. The traditionally applied default factors for subacute to semichronic (10), for semichronic to chronic (10), and for subacute to chronic exposure (100) corresponded with the 93, 87, and 99-percentiles of the respective distributions. Options are presented for a set of default values and probabilistic distributions for assessment factors for exposure duration based on data from the consulted historical database.     

Modelling effects of diquat under realistic exposure patterns in genetically differentiated populations of the gastropod Lymnaea stagnalis

Pesticide use leads to complex exposure and response patterns in non-target aquatic species, so that the analysis of data from standard toxicity tests may result in unrealistic risk forecasts. Developing models that are able to capture such complexity from toxicity test data is thus a crucial issue for pesticide risk assessment. In this study, freshwater snails from two genetically differentiated populations of Lymnaea stagnalis were exposed to repeated acute applications of environmentally realistic concentrations of the herbicide diquat, from the embryo to the adult stage. Hatching rate, embryonic development duration, juvenile mortality, feeding rate and age at first spawning were investigated during both exposure and recovery periods. Effects of diquat on mortality were analysed using a threshold hazard model accounting for time-varying herbicide concentrations. All endpoints were significantly impaired at diquat environmental concentrations in both populations. Snail evolutionary history had no significant impact on their sensitivity and responsiveness to diquat, whereas food acted as a modulating factor of toxicant-induced mortality. The time course of effects was adequately described by the model, which thus appears suitable to analyse long-term effects of complex exposure patterns based upon full life cycle experiment data. Obtained model outputs (e.g. no-effect concentrations) could be directly used for chemical risk assessment.     

Modelling population exposure-response functions for use in environmental risk assessment

Effective environmental management requires accurate prediction of the probable individual, population, and ecosystem responses associated with environmental hazards. While much is known about the short-term physiological impacts of toxicants at the individual level, little is known about the long-term responses of populations. This occurs, in part, because of the costs and difficulties associated with completing long-term studies. In the absence of such field data it is argued that modelling both bridges the existing information gap and provides a credible means of predicting long-term population responses. An individuals-based Atlantic salmon (Salmo salar) population dynamics model, adjusted to include laboratory-derived acute toxicity data, is used to measure recovery time in a population subjected to concentrations and durations of copper exposure characteristic of an accidental release of mine tailings. Selected recovery criteria are proposed and discussed in terms of their suitability for use in environmental risk assessment. The resulting model data are used to estimate population exposure-response functions and, for purposes of environmental risk assessment, to describe the cumulative probability distribution ofin-situ environmental damage. The output of the model suggests a recovery time of 15 to 20 years and significant increases in the variability of post-perturbation population levels.     

Setting Pesticide Reference Doses: A Retrospective Analysis Examining Key Data and Choices

Toxicity tests are widely used to set “acceptable” levels of chemical exposure. Different organizations have identified a base set of tests specifying a mix of endpoints, durations, and species to be tested. A specific test and endpoint is chosen as the basis for calculation of human health risk values like reference doses (RfDs). This study empirically evaluates the data and choices made in setting acute and chronic RfDs for 352 conventional pesticides. The results suggest that for Acute, Acute-Female Specific, and Chronic RfDs one test is used far more than others. Ninety-six percent of the 116 Acute Female-Specific RfDs relied on a developmental toxicity test and 78% of Chronic RfDs used the chronic bioassay. Tests in rats were used far more often than other species in all RfD calculations. For all types of RfDs a total uncertainty factor of 100 was most common although values as low as 1 and as high as 3000 were seen. These results provide insights not only into the science policy frameworks used, but also into ways toxicity testing and risk assessment may be streamlined and made more efficient.     

Challenges and Research Needs for Risk Assessment of Pesticides for Registration in Africa

Risk assessment is necessary for registration and risk management of new pesticides. The aim of this article is to discuss challenges that risk assessors in Africa face when conducting risk assessment of pesticides. Risk assessment requires toxicity assessment, environmental fate studies, and the use of models for occupational, dietary, residential, and environmental exposure assessments. Toxicity studies are very costly with the result that toxicity data used to register pesticides in Africa are often sourced from northern hemisphere countries. Assessors also often use exposure modeling results from the northern hemisphere. This is not an ideal approach as occupational exposure is influenced by agricultural practices, climatic conditions, and other factors. Furthermore, residential exposure models require time-location-activity information, exposure factors, and toxicokinetic rate constants for particular pesticides. Dietary exposure assessment needs accurate and comprehensive local food consumption data. Authorities in African countries should therefore generate the required data, despite these being very costly and tedious. Authorities should also provide guidance on the type of models and standard scenarios for estimating predicted environmental concentrations in various environmental compartments. It is recommended that higher educational institutions in Africa should incorporate risk assessment in general and pesticide toxicity and exposure models in particular in their curricula.     

Comparison of complex modeling strategies for prediction of a binary outcome based on a few,highly correlated predictors

Motivated by a clinical prediction problem, a simulation study was performed to compare different approaches for building risk prediction models. Robust prediction models for hospital survival in patients with acute heart failure were to be derived from three highly correlated blood parameters measured up to four times, with predictive ability having explicit priority over interpretability. Methods that relied only on the original predictors were compared with methods using an expanded predictor space including transformations and interactions. Predictors were simulated as transformations and combinations of multivariate normal variables which were fitted to the partly skewed and bimodally distributed original data in such a way that the simulated data mimicked the original covariate structure. Different penalized versions of logistic regression as well as random forests and generalized additive models were investigated using classical logistic regression as a benchmark. Their performance was assessed based on measures of predictive accuracy, model discrimination, and model calibration. Three different scenarios using different subsets of the original data with different numbers of observations and events per variable were investigated. In the investigated setting, where a risk prediction model should be based on a small set of highly correlated and interconnected predictors, Elastic Net and also Ridge logistic regression showed good performance compared to their competitors, while other methods did not lead to substantial improvements or even performed worse than standard logistic regression. Our work demonstrates how simulation studies that mimic relevant features of a specific data set can support the choice of a good modeling strategy.     

Development of Factors for Estimating Swimmers' Exposures to Chemicals in Swimming Pools

The U.S. Environmental Protection Agency (USEPA) has developed a model to estimate chemical-specific exposures to swimmers following pool treatments. This model is called SWIMODEL. The model was first introduced in 1996, and the USEPA has made refinements to the model over the last decade. One of the most significant parameters in this model is the duration of exposure. Unfortunately, there was no reliable source of data for these exposure durations, so highly conservative estimates were used in the earlier version of SWIMODEL. In order to address this data deficiency, a telephone survey of competitive swimming coaches was conducted to provide data for the refinement of the SWIMODEL. The survey collected data on practice durations and other factors that are potentially relevant to swimmer exposures. Two different surveys were developed. The first survey was designed for coaches of amateur swim club teams, and the second survey was for collegiate coaches. The survey found that the average practice durations for youth, adult Masters, and collegiate teams were 9.1, 6.4, and 17.0 hours/week, respectively. The use of these data to estimate acute, chronic, and lifetime exposures is discussed. The data developed in this study were used to update the exposure duration and frequency estimates in SWIMODEL for competitive swimmers based on survey responses. The survey was also used to determine the percentage of competitive swimmers that wear eye goggles and other protective gear, and to supplement other measurements of the incidental water ingestion of competitive swimmers.     

Changes in lung ATP concentration in the rat after low-level phosgene exposure

Inhibition of mitochondrial respiratory activity and decreased lung adenosine triphosphate (ATP) concentration occur following exposure to 240 ppm·min phosgene. To determine the relationship between energy stores and the onset of phosgene-induced pulmonary edema, we measured the ATP concentration in rapidly frozen rat lung tissue before and during pulmonary edema. Male Sprague-Dawley rats were exposed to phosgene for four hours at concentrations of 0.05 to 1.0 ppm (12, 30, 60,120, and 240 ppm·min). Lung wet and dry weight and ATP concentration were measured immediately after exposure and for three days postexposure. The accumulation of lavage fluid protein (LFP) was also measured as an index of damage or edema due to phosgene. Lung dry weight was significantly elevated one day postexposure to 0.5 ppm phosgene, while the LFP was elevated by 0.2 ppm phosgene. Time course studies at these doses of phosgene showed that decreased ATP levels preceded the onset of edema or increase in lung weight. The ATP values expressed on a per-lung basis showed that ATP levels were significantly lowered immediately following phosgene exposure, suggesting that the ATP changes were not the result of edema. This study is the first demonstration of a biochemical change that occurs following exposure to phosgene at a level significantly below the threshold limit value for this gas.     

Toxicity of γ-hexachlorocyclohexane (Lindane) to Gammarus pulex

SUMMARY. The acute toxicity of the pesticide γ-hexachlorocyclohexane (Lindane) to Gammarus pulex was determined by a standard procedure and by a modification of the conventional toxicity test protocol. Animals were exposed to a range of concentrations for various periods of time from 1 to 1000 min and transferred to clean water. Animals continued to die for up to 3 weeks after the initial exposure to the poison. Toxicity curves were obtained expressing the relationships between concentration of poison and median survival time (LT50) and between concentration of poison and median lethal exposure time, i.e., the duration of exposure required to cause the eventual death of half the animals. For a given concentration of pesticide, the duration of exposure which will cause 50% mortality is far less than is indicated by a conventional toxicity test in which animals are continuously exposed to the poison. It is suggested that the results of conventional acute lethal toxicity tests may have limited predictive value even when applied to field situations involving the discharge of lethal levels of pollutants to receiving waters.     

Cancer risk assessment for 1,3-butadiene: dose-response modeling from an epidemiological perspective

The dose-response assessment of the association between 1,3-butadiene (BD) and leukemia mortality among workers in the North American synthetic rubber industry is explored. Analyses are based on the most recent University of Alabama at Birmingham epidemiological study and exposure estimation. The U.S. EPA Science Advisory Board recommendations of using the most recent data and giving consideration to peak exposures to BD have been followed. If cumulative BD ppm-years is to be used as the predictor of the leukemia rate ratio, then the performance of that predictor is statistically significantly improved if the slope in the predictor is estimated with age and the cumulative number of BD peaks (where a BD peak is any exposure, regardless of duration, to a BD concentration above 100 ppm) added as categorical covariates. After age and the cumulative number of BD peaks are incorporated as categorical covariates in the Poisson regression model, the estimated concentration (EC(001)) corresponding to an excess risk of 0.001 as a result of continuous environmental exposure is 11.2 ppm; however, the estimated slope for BD cumulative ppm-years in the linear rate ratio for leukemia used to derive this EC(001) is not statistically significantly different from zero. Sensitivity analyses using alternative models indicate either essentially no risk or estimated EC(001) values of 9 and 77 ppm. Analyses suggesting the absence of a statistically significant low-dose risk versus cumulative BD ppm-years are presented. Sensitivity analyses of other malignant neoplasms of lymphatic and hematopoietic tissue (specifically, lymphoid and myeloid neoplasms) resulted in conclusions about the dose-response modeling methodology that were supportive of the methodology used for leukemia.     

U.S. EPA Reference Dose/Reference Concentration Methodology: Update on a Review of the Process

The U.S. Environmental Protection Agency (USEPA) has been reviewing several approaches to testing and risk assessment related to implementation of the Food Quality Protection Act (FQPA) and the Amendments to the Safe Drinking Water Act (SDWA), both signed into law in 1996. Based on recommendations from a review of issues related to children's health protection under these laws, the USEPA established the RfD Technical Panel to evaluate in depth the current reference dose (RfD) and reference concentration (RfC) process in general, and in particular with respect to how well children and other potentially sensitive subpopulations are protected. The RfD Technical Panel also was asked to consider scientific issues that have become of greater concern in RfD and RfC derivation (e.g., neurotoxicity, immunotoxicity), and to raise issues that should be explored or developed further for application in the RfD/RfC process. This paper provides the current status of the activities of the RfD Technical Panel. The Technical Panel has recommended that acute, short- term, and intermediate reference values should be set for chemicals, where possible, and that these values should be incorporated into the USEPA's Integrated Risk Information System (IRIS) Database. A review of current testing procedures is underway, including the endpoints assessed, life stages covered by exposure and outcome evaluation, and information that can be derived from current protocols on various durations of exposure. Data gaps identified for risk assessment include the types of pharmacokinetic data that should be collected, especially for developmental toxicity studies, the impact of aging on toxic responses occurring after early exposure as well as concomitant with exposure in old age, and information available on latency to response. The implications of the RfD Technical Panel's recommendations for various uncertainty factors are also being explored.     

Key issues for the development and application of the species sensitivity distribution (SSD) model for ecological risk assessment

The species sensitivity distribution (SSD) model is one of the most commonly used methods for ecological risk assessment based on the potentially affected fraction (PAF) of and the combined PAF (msPAF) as quantitative indicators. There are usually four steps for the development of SSD models and their applications: (1) obtain the toxicity data of the pollutants; (2) fit the SSD curves; (3) calculate the potentially affected fractions (PAFs) of the individual pollutants for the ecological risk assessment of an individual pollutant; and (4) calculate the accumulated multi-substance potentially affected fractions (msPAFs) for the joint ecological risk assessment of multiple pollutants. Among the above mentioned four steps, the first two steps are paramount. In the present study, the following six key issues are discussed: (1) how to select the appropriate species, (2) how to preprocess the toxicity data collected from the ecotoxicity database, (3) how to transform the acute toxicity data into chronic data, (4) how to best fit the toxicity data, (5) how to calculate the msPAF of multiple pollutants, and (6) how to determine the uncertainty of the SSD model”. In response to these questions, several principles were proposed to select appropriate species; three data processing methods, including the geometric mean, weight assigning and using all raw data without processing, were compared to determine the appropriate method for the DDT (dichloro diphenyl trichloroethane) toxicity data preprocessing. The method of acute to chronic ratio (ACR) and binary correlation analysis were contrasted using the zinc toxicity data for the transformation of the acute toxicity data into chronic data. The Burr III, Loglogistic and Lognormal models were compared to determine the best fit model using the DDT toxicity data for invertebrates. The concentration addition or response addition were discussed to calculate msPAF according to the toxic model of action (TMoA). The uncertainties of the SSD models for five heavy metals and for eight polycyclic aromatic hydrocarbons (PAHs) were performed. The comparison of the coefficients of variation (CVs) for the toxicity data and exposure levels in Lake Chaohu for eight polycyclic aromatic hydrocarbons (PAHs) were also presented to demonstrate the uncertainties of the ecological risks assessed by the SSD model based on 5000 Monte Carlo simulations.     

A Bayesian approach for joint modeling of cluster size and subunit-specific outcomes

In applications that involve clustered data, such as longitudinal studies and developmental toxicity experiments, the number of subunits within a cluster is often correlated with outcomes measured on the individual subunits. Analyses that ignore this dependency can produce biased inferences. This article proposes a Bayesian framework for jointly modeling cluster size and multiple categorical and continuous outcomes measured on each subunit. We use a continuation ratio probit model for the cluster size and underlying normal regression models for each of the subunit-specific outcomes. Dependency between cluster size and the different outcomes is accommodated through a latent variable structure. The form of the model facilitates posterior computation via a simple and computationally efficient Gibbs sampler. The approach is illustrated with an application to developmental toxicity data, and other applications, to joint modeling of longitudinal and event time data, are discussed.     

Use of in vivo genetic toxicity data for risk assessment

Mutagenicity studies have been used to identify specific agents as potential carconogens or other human health hazards; however, they have been used minimally for risk assessment or in determining permissible levels of human exposure. The poor predictive value of in vitro mutagenesis tests for carcinogenic activity and a lack of mechanistic understanding of the roles of mutagens in the induction of specific cancers have made these tests unattractive for the purpose of risk assessment. However, the limited resources available for carcinogen testing and large number of chemicals which need to be evaluated necessitate the incorporation of more efficient methods into the evaluation process. In vivo genetic toxicity testing can be recommended for this purpose because in vivo assays incorporate the metabolic activation pathways that are relevant to humans. We propose the use of a multiple end-point in vivo comprehensive testing protocol (CTP) using rodents. Studies using sub-acute exposure to low levels of test agents by routes consistent with human exposure can be a useful adjunct to methods currently used to provide data for risk assessment. Evaluations can include metabolic and pharmacokinetic endpoints, in addition to genetic toxicity studies, in order to provide a comprehensive examination of the mechanism of toxicity of the agent. A parallelogram approach can be used to estimate effects in non-accessible human tissues by using data from accessible human tissues and analogous tissues in animals. A categorical risk assessment procedure can be used which would consider, in order of priority, genetic damage in man, genetic damage in animals that is highly relevant to disease outcome (mutation, chromosome damage), and data from animals that is of less certain relevance to disease. Action levels of environmental exposure would be determined based on the lowest observed effect levels or the highest observed no effect levels, using sub-acute low level exposure studies in rodents. As an example, the known genotoxic effects of benzene exposure at low levels in man and animals are discussed. The lowest observed genotoxic effects were observed at about 1–10 parts per million for man and 0.04–0.1 parts per million in subacute animal studies. If genetic toxicity is to achieve a prominent role in evaluating carcinogens and characterizing germ-cell mutagens, minimal testing requirements must be established to ascertain the risk associated with environmental mutagen exposure. The use of the in vivo approach described here should provide the information needed to meet this goal. In addition, it should allow truly epigenetic or non-genotoxic carcinogens to be distinguished from the genotoxic carcinogens that are not detected by in vitro methods.     

Aquatic Risk Assessment of Metals in Sediment from South Florida Canals

A major watershed restoration effort is underway in south Florida, yet there are significant gaps in scientific information on exposure and risks of contaminants to its natural resources. We conducted a two-tier aquatic screening-level ecological risk assessment for metals that were monitored in sediment at 32 sampling sites in south Florida freshwater canals from 1990–2002. For tier 1, the chemicals (or metals) of potential ecological concern (COPECs) were identified as arsenic, cadmium, chromium, copper, lead, nickel and zinc based on their exceedences of Florida sediment quality guidelines at 10 sites. For tier 2, we used a probabilistic risk assessment method to compare distributions of predicted pore water exposure concentrations of seven metal COPECs with distributions of species response data from laboratory toxicity tests to quantify the likelihood of risk. The overlap of pore water concentrations (90th centile for exposure) for metal COPECs and the effects distributions for arthropods (10th centile of LC50s) and all species (10th centile of chronic NOECs) were used as a measure of potential acute and chronic risks, respectively. Arsenic (25%) in the Holey Land tracts, in Broward County north of Everglades National Park (ENP), and chromium (25%) in the C-111 freshwater system, at the east boundary of ENP, were the most frequently detected COPECs in sediment. Antimony (6%), zinc (6%) and lead (5%) were the least frequently detected COPECs in sediment. The 90th centile concentrations for bulk sediment were highest for zinc (at S-178) and lead (at S-176) in the C-111system. The 90th centile concentration for pore water exposure was highest for arsenic in the Holey Land tracts and lowest for cadmium and chromium. The estimated acute 10th centile concentration for effects was lowest for copper and arthropods. The probabilities of pore water exposures of copper exceeding the estimated acute 10th centile concentration from the species sensitivity distributions (SSD) of acute toxicity data (for arthropods) were 57 and 100% for copper at S-177 and S-178 in the C-111 system, respectively. The probability of pore water exposures of copper exceeding the estimated NOEC 10th centile concentration from the SSD of chronic toxicity data (for all species) was 93 and 100% for copper at S-177 and S-178, respectively. Uncertainties in exposure and effects analysis and risk characterization are identified and discussed. The study presents a straightforward approach to estimate exposure and potential risks of metals detected in sediment from south Florida canals.