Avis de l'Anses relatif à la Pertinence de la ré-évaluation de la VTR chronique par voie orale pour les ions perchlorate

Since 2011, ANSES has been asked on several occasions to examine the health risks associated with perchlorate (ClO4-) in DW following the identification of situations in which resources used for DW production were contaminated. In 2011, the Agency had recommended a guideline value (GV) for perchlorate in DW of 15 µg.L-1 for adult consumers, derived from the TRV of 0.7 µg.kg bw-1.d-1(ANSES, 2011), based on the inhibition of iodine uptake by the thyroid gland. ANSES had also advised against using water contaminated by perchlorate when preparing feeding bottles for infants aged up to 6 months, pending a national survey on perchlorate contamination of food, in particular to clarify the concentrations found in the powdered infant formula used for bottle preparation.In 2012, in its opinion concerning epidemiological studies on associations between exposure to perchlorate in DW and thyroid function in specific populations, the Agency had concluded that: "the results from the evaluated epidemiological studies do not enable conclusions to be drawn concerning the possible association between thyroid-stimulating hormone (TSH) levels and perchlorate concentrations in drinking water in pregnant women and newborns [...]. The absence of information concerning the iodine status of the studied populations makes it difficult to interpret the published epidemiological data." (ANSES, 2012).In its opinion of 8 April 2014 on the presence of perchlorate in infant formula and in DW, the Agency had noted the same limitations relating to the available epidemiological studies and insisted on the need to take account of the iodine status of the study population for assessing the health impact of human exposure to perchlorate and for interpreting the published epidemiological data.In its opinion of 26 December 2018, the Agency had concluded that recent epidemiological studies, including the one by the French Institute for Public Health Surveillance1(InVS) published in 2016, did not provide any additional conclusive evidence on the biological or clinical effects of perchlorate versus those taken into account in the previous ANSES opinions (ANSES 2011, 2012, 2014). The data examined in this opinion did not call into question the conclusions of the previous ANSES opinions concerning the hazard characterisation and the RV proposed by the Agency in 2011.However, with regard to the assessment of exposure to perchlorate, new data made it possible to estimate the average oral exposure of the adult population to perchlorate. This estimate was based firstly on data on perchlorate contamination of food collected by the Directorate General for Competition Policy, Consumer Affairs and Fraud Control (DGCCRF) as part of its national surveys, and secondly on data on DW contamination for the period 2014-2017 produced by the Regional Health Agencies (ARSs) and available in the SISE-Eaux database. On this basis, ANSES estimated that exposure via DW consumption accounted for about 25% of ingestion exposure. As this estimate was in line with the data in the literature and was close to the 20% default percentage defined by the World Health Organization (WHO) in 2016, the Agency felt it necessary to lower to 20% the share of exposure via water (which had previously been set at 60%) used for calculating the GV of perchlorate in DW for adults. As a result, the proposed GV was lowered to 5 µg.L-1 for the adult population.ANSES also emphasised that the existing data on food contamination were too incomplete to characterise the distribution of the general population's exposure to perchlorate by the oral route and to assess the health risk. It therefore recommended that perchlorate be included in the third French Total Diet Study (TDS3).In the absence of new data on the contamination of infant formula with perchlorate since the 2014 expert appraisal, the Agency also reiterated the conclusions of this earlier expert appraisal: "daily intakes of perchlorate, calculated on the basis of perchlorate levels in infant formulas available on the French market, do not exceed the TRV of 0.7 µg.kg bw-1.d-1for 95% of the population of children aged under 6 months consuming infant formula based on an average concentration of perchlorate of 1 µg.L-1 in DW for reconstituting infant bottles" (ANSES, 2018).Lastly, ANSES indicated in this same 2018 opinion, in Section 3.2.8 concerning the conclusions on the choice of the TRV, that the Working Group on "Assessment of the health risks associated with chemical parameters in drinking water" (ERS EDCH) and the Expert Committee (CES) on "Water" considered that in any future assessment of the health risks associated with the ingestion of perchlorate, following publication of the work under way by the US EPA, it would be necessary to re-examine the method for determining the critical dose and establishing the TRV for perchlorate."All the agencies establishing chronic oral TRVs for perchlorate have chosen inhibition of iodine uptake by the thyroid as the critical effect (Table 1 on page 8).In 2019, the US EPA adopted a different approach based on toxicokinetic-toxicodynamic modelling to assess the impact of perchlorate exposure on thyroid hormone (TH) production and the effect of a reduction in maternal plasma free thyroxine (fT4) concentrations during the first trimester of pregnancy on the child's neurodevelopment, as measured by a decrease in intelligence quotient. The US EPA suggested adopting a TRV of 2.2 µg.kg bw-1.d-1, resulting in a maximum contaminant level of 56 µg.L-1 perchlorate in DW.As mentioned previously, and in view of the management difficulties encountered, the DGS had formally asked ANSES to analyse the US EPA's work on assessing the health risks associated with the presence of perchlorate in DW, in order to provide the ARSs with appropriate and proportionate guidance on the health risk.In view of this, ANSES was asked to re-examine the assessment of the health risks associated with the ingestion of perchlorate, in light of the US EPA's work. The first deliverable of this formal request therefore had the ultimate objective of determining whether the ANSES TRV of 0.7 µg.kg bw1.d-1 should be maintained. This first deliverable concluded that this TRV should not be called into question in light of the US EPA's work, mainly because of major uncertainties about the predictive ability of the model used. The Agency therefore decided to retain its chronic oral TRV established in 2011, based on the no-effect level observed in the study by Greer et al. (2002). However, all the TRVs established since that date are based on a Benchmark Dose (BMD) approach, so the Agency therefore issued an internal request to assess the relevance of this approach, based in particular on recent publications (Weterings et al. 2016, Bruce et al. 2018, Haber et al. 2021).

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.