Is the cholesterol-perfluoroalkyl substance association confounded by dietary fiber intake?: a Bayesian analysis of NHANES data with adjustment for measurement error in fiber intake.

BACKGROUND: Serum concentrations of total cholesterol and related lipid measures have been associated with serum concentrations of per- and polyfluoroalkyl substances (PFAS) in humans, even among those with only background-level exposure to PFAS. Fiber is known to decrease serum cholesterol and a recent report based on National Health and Nutrition Examination Survey (NHANES) showed that PFAS and fiber are inversely associated. We hypothesized that confounding by dietary fiber may account for some of the association between cholesterol and PFAS. METHODS: We implemented a Bayesian correction for measurement error in estimated intake of dietary fiber to evaluate whether fiber confounds the cholesterol-PFAS association. The NHANES measure of diet, two 24-h recalls, allowed calculation of an estimate of the "true" long-term fiber intake for each subject. We fit models to the NHANES data on serum cholesterol and serum concentration of perfluorooctanoic acid (PFOA) and two other PFAS for 7,242 participants in NHANES. RESULTS: The Bayesian model, after adjustment for soluble fiber intake, suggested a decrease in the size of the coefficient for PFOA by 6.4% compared with the fiber-unadjusted model. CONCLUSIONS: The results indicated that the association of serum cholesterol with PFAS was not substantially confounded by fiber intake.

Bayesian hierarchical evaluation of dose-response for peanut allergy in clinical trial screening.

Risk-based labeling based on the minimal eliciting doses (EDs) in sensitized populations is a potential replacement for precautionary allergen labeling of food allergens. We estimated the dose-response distribution for peanut allergen using data from double-blind placebo-controlled food challenges (DBPCFCs) conducted in the US at multiple sites, testing a population believed to be similar to the general U.S. food allergic population. Our final (placebo-adjusted) dataset included 548 challenges of 481 subjects. Bayesian hierarchical analysis facilitated model fitting, and accounted for variability associated with various levels of data organization. The data are best described using a complex hierarchical structure that accounts for inter-individual variability and variability across study locations or substudies. Bayesian model averaging could simultaneously consider the fit of multiple models, but the Weibull model dominated so strongly that model averaging was not needed. The ED01 and ED05 (and 95% credible intervals) are 0.052 (0.021, 0.13) and 0.49 (0.22, 0.97) mg peanut protein, respectively. Accounting for challenges with severe reactions at the LOAEL, by using the dose prior to the LOAEL as the new LOAEL, the ED01 drops to 0.029 (0.014, 0.074) mg peanut protein. Our results could aid in establishing improved food labeling guidelines in the management of food allergies.

Impact of updated BMD modeling methods on perchlorate and chlorate assessments of human health hazard.

Several exposure limits for perchlorate have been developed based on an early key event, inhibition of radioactive iodide uptake (RAIU) by the thyroid. These assessments have used a variety of definitions of the point of departure. The current assessment revisited the modeling for inhibition of RAIU, using state of the science methods. Bayesian hierarchical modeling was used to account for the repeated measures on the same individuals in the key dataset, and the underlying Beta distribution used for the modeling correctly reflected the bounding of RAIU between 0 and 1. We defined the BMR as a point value of 8% RAIU (rather than a change in RAIU), based on descriptions in the medical literature that RAIU below this value is considered abnormal. Because a definition of the BMR based on the mean response would correspond to about 50% of the population with a response below the BMR at the benchmark dose, we used a hybrid definition of the BMR. That is, the BMD was defined as the dose at which it was estimated that there would be a 10% extra risk in the population of having RAIU of 8% or lower. The resulting point of departure based on the BMDL was 0.03 mg/kg-day.

Quantitative Risk Assessment: Developing a Bayesian Approach to Dichotomous Dose-Response Uncertainty.

Model averaging for dichotomous dose-response estimation is preferred to estimate the benchmark dose (BMD) from a single model, but challenges remain regarding implementing these methods for general analyses before model averaging is feasible to use in many risk assessment applications, and there is little work on Bayesian methods that include informative prior information for both the models and the parameters of the constituent models. This article introduces a novel approach that addresses many of the challenges seen while providing a fully Bayesian framework. Furthermore, in contrast to methods that use Monte Carlo Markov Chain, we approximate the posterior density using maximum a posteriori estimation. The approximation allows for an accurate and reproducible estimate while maintaining the speed of maximum likelihood, which is crucial in many applications such as processing massive high throughput data sets. We assess this method by applying it to empirical laboratory dose-response data and measuring the coverage of confidence limits for the BMD. We compare the coverage of this method to that of other approaches using the same set of models. Through the simulation study, the method is shown to be markedly superior to the traditional approach of selecting a single preferred model (e.g., from the U.S. EPA BMD software) for the analysis of dichotomous data and is comparable or superior to the other approaches.

Pharmacokinetic bias analysis of an association between clinical thyroid disease and two perfluoroalkyl substances.

Exposure to perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS) has been associated with the occurrence of thyroid disease in some epidemiologic studies. We hypothesized that in a specific epidemiologic study based on the National Health and Nutrition Examination Survey, the association of clinical thyroid disease with serum concentration of PFOA and PFOS was due to reverse causality. Thyroid hormone affects glomerular filtration, which in turn affects excretion of PFOA and PFOS. We evaluated this by linking a model of thyroid disease status over the lifetime to a physiologically based pharmacokinetic model of PFOA and PFOS. Using Monte Carlo methods, we simulated the target study population and analyzed the data using multivariable logistic regression. The target and simulated populations were similar with respect to age, estimated glomerular filtration rate, serum concentrations of PFOA and PFOS, and prevalence of clinical thyroid disease. The analysis showed little or no evidence of bias from the hypothesized mechanism. The largest bias was for the fourth quartile of PFOA in females, with an odds ratio of 0.93 (95% CI, 0.90, 0.97). The reported odds ratio of clinical thyroid disease for this group was 1.63 (1.07, 2.47), and if it were corrected for the bias would have been 1.74 (1.14, 2.65). Our results suggest that little of the reported association in the target study was due to reverse causality.

Quantitative bias analysis of the association between subclinical thyroid disease and two perfluoroalkyl substances in a single study.

Exposure to perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS) has been associated with the occurrence of thyroid disease in some epidemiologic studies. We hypothesized that in a specific epidemiologic study based on the National Health and Nutrition Examination Survey, the association of subclinical thyroid disease with serum concentration of PFOA and PFOS was due to reverse causality. Thyroid hormone affects glomerular filtration, which in turn affects excretion of PFOA and PFOS. We evaluated this by linking a model of thyroid disease status over the lifetime to physiologically based pharmacokinetic models of PFOA and PFOS. Using Monte Carlo methods, we simulated the target study population and analyzed the data using multivariable logistic regression. The target and simulated populations were similar with respect to age, estimated glomerular filtration rate, serum concentrations of PFOA and PFOS, and prevalence of subclinical thyroid disease. Our findings suggest that in the target study the associations with subclinical hypothyroidism were overstated and the results for subclinical hyperthyroidism were, in general, understated. For example, for subclinical hypothyroidism in men, the reported odds ratio per ln(PFOS) increase was 1.98 (95% CI 1.19-3.28), whereas in the simulated data the bias due to reverse causality gave an odds ratio of 1.19 (1.16-1.23). Our results provide evidence of bias due to reverse causality in a specific cross-sectional study of subclinical thyroid disease with exposure to PFOA and PFOS among adults.

A model of functional thyroid disease status over the lifetime.

Mathematical models of the natural history of disease can predict incidence rates based on prevalence data and support simulations of populations where thyroid function affects other aspects of physiology. We developed a Markov chain model of functional thyroid disease status over the lifetime. Subjects were in one of seven thyroid disease states at any given point in their lives [normal, subclinical hypothyroidism, overt hypothyroidism, treated thyroid disease (ever), subclinical hyperthyroidism, overt hyperthyroidism, and reverted to normal thyroid status]. We used a Bayesian approach to fitting model parameters. A priori probabilities of changing from each disease state to another per unit time were based on published data and summarized using meta-analysis, when possible. The probabilities of changing state were fitted to observed prevalence data based on the National Health and Nutrition Examination Survey 2007-2012. The fitted model provided a satisfactory fit to the observed prevalence data for each disease state, by sex and decade of age. For example, for males 50-59 years old, the observed prevalence of ever having treated thyroid disease was 4.4% and the predicted value was 4.6%. Comparing the incidence rates of treated disease predicted from our model with published values revealed that 82% were within a 4-fold difference. The differences seemed to be systematic and were consistent with expectation based on national iodine intakes. The model provided new and comprehensive estimates of functional thyroid disease incidence rates for the U.S. Because the model provides a reasonable fit to national prevalence data and predicts thyroid disease status over the lifetime, it is suitable for simulating populations, thereby making possible quantitative bias analyses of selected epidemiologic data reporting an association of thyroid disease with serum concentrations of environmental contaminants.

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.

Derivation of an oral toxicity reference value for nickel.

Nickel (Ni) is in the earth's crust and can be found in environmental compartments such as water, soil, and air, as well as food. This paper presents an assessment of the oral nickel toxicity data in support of non-cancer health-based oral exposure limits or toxicity reference values (TRVs). This paper derives TRVs for three populations of interest: adults, toddlers, and people who have been dermally sensitized to nickel. The adult/lifetime TRV of 20 µg Ni/kg-day is based on post-implantation loss/perinatal mortality in a 2-generation reproductive study in rats. Several recent assessments by regulatory agencies have used the same study and endpoint, but the dose-response modeling conducted here was more appropriate for the study design. Toxicokinetic data from rats and humans indicate that the applied uncertainty factors are very conservative. Because the endpoint relates to fetal exposure and is not relevant to toddlers, a toddler TRV was derived based on decreased body weight in young rats; this TRV was also 20 µg Ni/kg-day. A separate TRV of 4 µg Ni/kg in addition to Ni in food was derived for protection of nickel-sensitized populations from flare-up of dermatitis, based on studies of single exposures in humans under conditions that maximize oral absorption.

Evaluation of Serum Lipid, Thyroid, and Hepatic Clinical Chemistries in Association With Serum Perfluorooctanesulfonate (PFOS) in Cynomolgus Monkeys After Oral Dosing With Potassium PFOS.

An oral dose study with perfluorooctanesulfonate (PFOS) was undertaken to identify potential associations between serum PFOS and changes in serum clinical chemistry parameters in purpose-bred young adult cynomolgus monkeys (Macaca fascicularis). In this study, control group (n = 6/sex) was sham-dosed with vehicle (0.5% Tween 20 and 5% ethanol in water), low-dose group (n = 6/sex) received 1 single K+PFOS dose (9 mg/kg), and high-dose group (n = 4-6/sex) received 3 separate K+ PFOS doses (11-17.2 mg/kg). Monkeys were given routine checkups and observed carefully for health problems on a daily basis. Scheduled blood samples were drawn from all monkeys prior to, during, and after K+PFOS administration for up to 1 year and they were analyzed for PFOS concentrations and clinical chemistry markers for coagulation, lipids, hepatic, renal, electrolytes, and thyroid-related hormones. No mortality occurred during the study. All the monkeys were healthy, gained weight, and were released back to the colony at the end of the study. The highest serum PFOS achieved was approximately 165 µg/ml. When compared with time-matched controls, administration of K+PFOS to monkeys did not result in any toxicologically meaningful or clinically relevant changes in serum clinical measurements for coagulation, lipids, hepatic, renal, electrolytes, and thyroid-related hormones. A slight reduction in serum cholesterol (primarily the high-density lipoprotein fraction), although not toxicologically significant, was observed. The corresponding lower-bound fifth percentile benchmark concentrations (BMCL1sd) were 74 and 76 µg/ml for male and female monkeys, respectively. Compared to the 2013-2014 geometric mean serum PFOS level of 4.99 ng/ml (0.00499 µg/ml) in US general population reported by CDC NHANES, this represents 4 orders of magnitude for margin of exposure.