Ann Parker

Evaluation of concentration―response options for diacetyl in support of occupational risk assessment

The current emphasis on occupational exposures to diacetyl has led to new research on its effects. We evaluated whether the data are sufficient to support a transition from a hazard-based risk management approach to a quantitative occupational risk assessment approach, characterized by developing a health-based occupational exposure limit (OEL). Inhalation health effects data were evaluated and issues and uncertainties related to occupational risk assessment needs were identified. A systematic hazard characterization, supported by both the toxicology and epidemiology literature, showed that the respiratory tract effects of diacetyl are the primary end points of relevance for developing an OEL. In an effort to provide a systematic approach for the analysis of the issues that need to be considered in developing an occupational risk assessment for diacetyl, a potential OEL was derived. A concentration-response assessment was completed using tracheobronchial effects in mice as the critical effect. The resulting benchmark concentration (lower bound estimate or BMCL) was adjusted to a human equivalent concentration of 1.8 ppm. A composite uncertainty factor of 10 was recommended to account for extrapolation from an adjusted BMCL from an animal study and for human variability in sensitivity and taking into account other uncertainties in the overall database. The resulting OEL recommendation of 0.2 ppm as a time-weighted average (TWA) was supported by the current occupational epidemiology literature. This evaluation showed that a health-based OEL value can be derived for diacetyl with moderate to high confidence.

Proposal of new uncertainty factor application to derive tolerable daily intake

We propose new uncertainty factors (UFs) and a new subdivision of default factors in chemical risk assessment using a probabilistic approach based on the latest applicable information. Rounded values of 150 for mice, 100 for hamsters and rats, and 40 for rabbits, monkeys and dogs for inter- and intra-species differences (UF(AH)) were derived from the probabilistic combination of two log-normal distributions. Further calculation of additional UFs when chronic data (UF(S)) or NOAEL (UF(L)) are lacking was conducted using available log-normal distribution information. The alternative UF(S) and UF(L) values of 4 are considered to be appropriate for both cases where data are lacking. The default contributions of inter-species difference (UF(A)) and intra-species difference (UF(H)) to the UF(AH) of 100 for hamsters and rats as an example are considered to be 25 and 4, respectively. The UF(A) of 25 was subdivided into 25(0.6) (i.e., 7.0) for pharmacokinetics (PK) (UF(A,PK)) and 25(0.4) (i.e., 3.6) for pharmacodynamics (PD) (UF(A,PD)), and the UF(H) of 4 was evenly subdivided into 4(0.5) (i.e., 2) (UF(H,PK) and UF(H,PD)), to account for chemical-specific difference data between humans and laboratory animals for PK and/or PD. These default UFs, which come from actual experimental data, may be more appropriate than previous default UFs to derive tolerable daily intake values.

Past and Future Use of Default Assumptions and Uncertainty Factors: Default Assumptions, Misunderstandings, and New Concepts

. (1996) show examples from the USEPA and Health Canada where data havebeen used to support the selection of uncertainty factors values other than default.Manyoftheunderlyingassumptionsandchoicesofuncertaintyfactorsinnoncancerrisk assessment are similar among these health agencies worldwide. Such similarityis one foundation of global harmonization of risk assessment methods, a formidabletaskundertakenbytheInternationalProgrammeonChemicalSafety(Sonich-Mullin1995; IPCS 1994, 2001, 2005).Uncertainty factors are not precise (Felter and Dourson 1998). The underlyingdata that support these factors demonstrate large variability. However, the incorpo-rationofallavailablescientificdataintotheriskassessmentprocessfostersincreasedresearchandultimatelyreducesuncertainty.This,inturn,supportstheuseofChem-ical Specific Adjustment Factors (CSAF), resulting in noncancer risk assessments inwhich greater confidence can be placed. Examples where research has reduced theuse of default factors are readily found (

Derived Reference Doses (RfDs) for the environmental degradates of the herbicides alachlor and acetochlor: Results of an independent expert panel deliberation

An independent peer expert panel was convened under the auspices of the Alliance for Risk Assessment (ARA) to review toxicology data and derive oral Reference Doses (RfDs) for four environmental degradates of the acetanilide herbicides, alachlor and acetochlor. The degradates included in this evaluation were (1) alachlor tertiary-ethanesulfonic acid (ESA), (2) alachlor tertiary-oxanilic acid (OXA), (3) acetochlor ESA, and (4) acetochlor OXA. Each degradate was judged to have sufficient data for developing low to medium confidence RfD, with use of an additional uncertainty factor (UF) to cover data gaps. Body weight decreases were identified as the most sensitive treatment-related adverse effect for RfD development. A composite UF of 1000 (10 for human variability in sensitivity, 10 for interspecies differences in sensitivity, and 10 for subchronic to chronic and database deficiency combined; i.e., 10(A)x10(H)x10(S&D)) for each degradate was considered reasonable, while noting that an argument could be made for an UF of 3000 (10(A)x10(H)x30(S&D)). Based on the available data, an oral RfD of 0.2 mg/kg-day is recommended for both acetochlor ESA and acetochlor OXA and an oral RfD of 0.8 mg/kg-day is recommended for both alachlor ESA and alachlor OXA.

Mode of action analysis for liver tumors from oral 1,4-dioxane exposures and evidence-based dose response assessment

1,4-Dioxane is found in consumer products and is used as a solvent in manufacturing. Studies in rodents show liver tumors to be consistently reported after chronic oral exposure. However, there were differences in the reporting of non-neoplastic lesions in the livers of rats and mice. In order to clarify these differences, a reread of mouse liver slides from the 1978 NCI bioassay on 1,4-dioxane in drinking water was conducted. This reread clearly identified dose-related non-neoplastic changes in the liver; specifically, a dose-related increase in the hypertrophic response of hepatocytes, followed by necrosis, inflammation and hyperplastic hepatocellular foci. 1,4-Dioxane does not cause point mutations, DNA repair, or initiation. However, it appears to promote tumors and stimulate DNA synthesis. Using EPA Guidelines (2005), the weight of the evidence suggests that 1,4-dioxane causes liver tumors in rats and mice through cytotoxicity followed by regenerative hyperplasia. Specific key events in this mode of action are identified. A Reference Dose (RfD) of 0.05mg/kgday is proposed to protect against regenerative liver hyperplasia based on a benchmark dose (BMD) approach. Based on this RfD, a maximum contaminant level goal of 350μg/L is proposed using a default relative source contribution for water of 20%.

Safety assessment of boron by application of new uncertainty factors and their subdivision

The available toxicity information for boron was reevaluated and four appropriate toxicity studies were selected in order to derive a tolerable daily intake (TDI) using newly proposed uncertainty factors (UFs) presented in Hasegawa et al. (2010). No observed adverse effect levels (NOAELs) of 17.5 and 8.8 mgB/kg/day for the critical effect of testicular toxicity were found in 2-year rat and dog feeding studies. Also, the 95% lower confidence limit of the benchmark doses for 5% reduction of fetal body weight (BMDL(05)) was calculated as 44.9 and 10.3 mgB/kg/day in mouse and rat developmental toxicity studies, respectively. Measured values available for differences in boron clearance between rats and humans and variability in the glomerular filtration rate (GFR) in pregnant women were used to derive chemical specific UFs. For the remaining uncertainty, newly proposed default UFs, which were derived from the latest applicable information with a probabilistic approach, and their subdivided factors for toxicokinetic and toxicodynamic variability were applied. Finally, overall UFs were calculated as 68 for rat testicular toxicity, 40 for dog testicular toxicity, 247 for mouse developmental toxicity and 78 for rat developmental toxicity. It is concluded that 0.13 mgB/kg/day is the most appropriate TDI for boron, based on rat developmental toxicity.

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

Abstract 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.

A tiered asthma hazard characterization and exposure assessment approach for evaluation of consumer product ingredients

Asthma is a complex syndrome with significant consequences for those affected. The number of individuals affected is growing, although the reasons for the increase are uncertain. Ensuring the effective management of potential exposures follows from substantial evidence that exposure to some chemicals can increase the likelihood of asthma responses. We have developed a safety assessment approach tailored to the screening of asthma risks from residential consumer product ingredients as a proactive risk management tool. Several key features of the proposed approach advance the assessment resources often used for asthma issues. First, a quantitative health benchmark for asthma or related endpoints (irritation and sensitization) is provided that extends qualitative hazard classification methods. Second, a parallel structure is employed to include dose-response methods for asthma endpoints and methods for scenario specific exposure estimation. The two parallel tracks are integrated in a risk characterization step. Third, a tiered assessment structure is provided to accommodate different amounts of data for both the dose-response assessment (i.e., use of existing benchmarks, hazard banding, or the threshold of toxicological concern) and exposure estimation (i.e., use of empirical data, model estimates, or exposure categories). Tools building from traditional methods and resources have been adapted to address specific issues pertinent to asthma toxicology (e.g., mode-of-action and dose-response features) and the nature of residential consumer product use scenarios (e.g., product use patterns and exposure durations). A case study for acetic acid as used in various sentinel products and residential cleaning scenarios was developed to test the safety assessment methodology. In particular, the results were used to refine and verify relationships among tiered approaches such that each lower data tier in the approach provides a similar or greater margin of safety for a given scenario.

Application of Markov chain Monte Carlo analysis to biomathematical modeling of respirable dust in US and UK coal miners.

A biomathematical model was previously developed to describe the long-term clearance and retention of particles in the lungs of coal miners. The model structure was evaluated and parameters were estimated in two data sets, one from the United States and one from the United Kingdom. The three-compartment model structure consists of deposition of inhaled particles in the alveolar region, competing processes of either clearance from the alveolar region or translocation to the lung interstitial region, and very slow, irreversible sequestration of interstitialized material in the lung-associated lymph nodes. Point estimates of model parameter values were estimated separately for the two data sets. In the current effort, Bayesian population analysis using Markov chain Monte Carlo simulation was used to recalibrate the model while improving assessments of parameter variability and uncertainty. When model parameters were calibrated simultaneously to the two data sets, agreement between the derived parameters for the two groups was very good, and the central tendency values were similar to those derived from the deterministic approach. These findings are relevant to the proposed update of the ICRP human respiratory tract model with revisions to the alveolar-interstitial region based on this long-term particle clearance and retention model.

Cleaning and asthma: A systematic review and approach for effective safety assessment

ABSTRACT Research indicates a correlative relationship between asthma and use of consumer cleaning products. We conduct a systematic review of epidemiological literature on persons who use or are exposed to cleaning products, both in occupational and domestic settings, and risk of asthma or asthma‐like symptoms to improve understanding of the causal relationship between exposure and asthma. A scoring method for assessing study reliability is presented. Although research indicates an association between asthma and the use of cleaning products, no study robustly investigates exposure to cleaning products or ingredients along with asthma risk. This limits determination of causal relationships between asthma and specific products or ingredients in chemical safety assessment. These limitations, and a lack of robust animal models for toxicological assessment of asthma, create the need for a weight‐of‐evidence (WoE) approach to examine an ingredient or products asthmatic potential. This proposed WoE method organizes diverse lines of data (i.e., asthma, sensitization, and irritation information) through a systematic, hierarchical framework that provides qualitatively categorized conclusions using hazard bands to predict a specific product or ingredients potential for asthma induction. This work provides a method for prioritizing chemicals as a first step for quantitative and scenario‐specific safety assessments based on their potential for inducing asthmatic effects. Acetic acid is used as a case study to test this framework. HighlightsAn updated systematic review of epidemiology studies leaves unresolved the basis for causal links with asthma.A data integration and hazard characterization system for asthma risk assessment is presented.The approach is demonstrated with an acetic acid ingredient case study.