(511c) Risk Assessment of Phthalates and Dinch Starting from Human Biomonitoring Data

Phthalate esters are widely used industrial chemicals. Their major portion, found in the environment, is the result of the slow release of phthalates from plastics and other phthalate containing articles due to weathering. A set of realistic exposure scenarios were evaluated for the phthalates, including assessments that start from environmental emissions to environmental and micro-environmental concentrations, food items residues, concentration in dust and in soil, concentration in ambient and indoor air, as well as exposure to consumer products. Phthalates exposure is associated to various pathways and routes, which are defined by their intended uses and physicochemical properties. To better validate the integrated exposure models, human biomonitoring data (HBM) were used. The idea was to run forward the integrated exposure models and to validate the expected relevant biomarker levels (e.g. urinary metabolites of phthalates) against the existing real life HBM data. Although this modelling exercise would be facilitated by individual HBM data, it was accompanied by detailed exposure related information (microenvironments encountered, dietary habits, frequency of use of consumer products). Human biomonitoring offers unique opportunities in modern risk assessment, on condition that the biomonitoring data are properly interpreted, especially within the regulatory context. A major advantage of HBM data is that they provide an integrated overview of the body burden to xenobiotics that an individual is exposed to; hence they serve as a very good approximation of aggregate exposure.

For modelling phthalate compounds, the INTEGRA modelling platform was used. The overall modeling framework for assessing the source to dose continuum with regard to DEHP, was built upon a unified simulation environment (asclxtreme), comprised of four interconnected modules, namely: multimedia indoor air quality module, exposure assessment modeling, internal dose, uncertainty and variability across all stages of the assessment. The generic PBTK model developed in INTEGRA is designed to describe the ADME processes occurring in the human body at different life stages in as much as possible detail, so as to be easily applicable to a broad variety of chemicals after proper parameterization. The model in its generic form includes the parent compound and up to three generations of potential metabolites. Advanced QSAR models are used to estimate physicochemical and biochemical parameters of the model in order to expand its applicability domain to a large chemical space. Environmental and exposure data from several studies, collected in the framework of HBM4EU consortium, were fed into INTEGRA assuming multi-pathway exposure including inhalation (both gaseous phase and particles), dietary ingestion through several food items and eventually emissions from building materials and use of consumer products.

In most of the studies, the daily intake of DEHP, DiNP and DnBP was close or above 1 μg/kg_bw/d, while for DiNCH and especialy for BBzP daily intake seemed to be one order of magnitude lower. For DEHP case study, the estimated concentrations for the several media were 1.5 μg/m³ for the gaseous phase, 5.5 μg/m³ for the particles and 3600 μg/gr for settled dust. These levels of predicted urinary levels of DEHP metabolites that fluctuate around 11 μg/L for the sum of the three major metabolites (MEHP, 5-OH-MEHP and 5-oxo-MEHP), are within the same magnitude of order of the measured levels that are within 10 to 100 μg/L. Based on the above, exposure to DEHP from all pathways and routes exceeds 10 μg/kg_bw/d, especially for neonates, where dust ingestion seems to be a significant contributor to the overall intake. For phthalates, daily intake estimates are usually one or two orders of magnitude below the respective TDI, with the exception of BBzP, for which intake estimates of the upper part of the exposure distribution is close to the threshold of 10 μg/kg_bw/d. Daily intake estimates for phthalates were derived based on the exposure reconstruction results of the available aggregate HBM data.

Also, the assessment of internal dose in different biological tissues and fluids of the phthalates based on the combination of external exposure determinants/modifiers and HBM measurements was made. A key point towards this translation is the use of physiology based toxicokinetic models, that allow for the translation of complex and highly dynamic multi-route exposure regimes, into time dynamic internal concentration in human tissues. Towards this direction, exposure assessment for running forward the PBTK models was necessary. These data came either from bottom up approaches, namely starting from environmental releases, environmental media and food item contamination, consumer product use, or by intake data that have been estimated starting from reconstruction of HBM data. The type of the used approach, relies on the relative availability of data from both directions, as well as the complexity of exposure pathways and routes that relate to the specific chemical. For estimating the internal dose of phthalates, exposure to phthalates (in terms of daily intake in μg/kg_bw/d) have been estimated through exposure reconstruction of human biomonitoring data from the relevant cohorts. Based on the availability of data, intake estimates and the respective internal dose expressed as concentration of the parent compound in blood has been estimated for DEHP, BBzP, DiNP, DnBP and DiNCH. For all compounds studied internal concentration was in the range of ng/L.

Overall, the use of expected biomonitored levels is a very useful approach for assessing the validity of the overall exposure assessment, because biomonitoring includes the contribution from all sources, pathways and routes.