Determination of a guidance value for the communication of individual-level biomonitoring data for urinary arsenic
Introduction
Biomonitoring, which involves measuring contaminants concentrations in the human body through biomarkers, has become a commonly tool used in populational survey. The main goal of such biomonitoring surveys is to assess a population's environmental exposure to contaminants and to support subsequent health risk management (Angerer et al., 2007; Haines et al., 2011). Inorganic arsenic (iAs), a recognized human carcinogen, is often included in biomonitoring programs (Angerer et al., 2011; CDC, 2021; Health Canada, 2019; IARC, 2012).
Populations are exposed to iAs mainly from drinking water (DW) and some types of foods (e.g. rice, rice-based products, other grains and grain-based products and certain fruit juices and vegetables) (ATSDR, 2007; European Food Safety Authority (EFSA) et al., 2021; IARC, 2012; Joint F.A.O./WHO Expert Committee on Food Additives, 2011). The biomarker most commonly used to determine exposure to iAs is the sum of urinary inorganic-related arsenic species (SUIAS), which corresponds to the sum of the inorganic forms As (III) and As (V) and their metabolites DMA and MMA (Hsueh et al., 2002). In a health risk assessment perspective, the analysis of the SUIAS is considered more useful than measurements of total urinary arsenic. Indeed, this latter biomarker is strongly influenced by the presence of the much less toxic organic forms of As (eg: arsenobetaine ans arsenosugars) from the consumption of seafood (fish, shellfish and algae) (CDC, 2017; Navas-Acien et al., 2011).
For many biomarkers of toxicant exposure such as SUIAS, the paucity of available guidance values strongly challenges the communication to participants of biomonitoring campaigns of their individual results (Haines et al., 2011). This barrier is even more limitating in the case of iAs, in particular because the SUIAS data show a high interindividual variability and due to the premise of absence of threshold exposure for its carcinogenic effect (Concha et al., 2006; Health Canada, 2006; IARC, 2012; Lindberg et al., 2006). It results in missed opportunities to make potentially overexposed participants aware of their specific situation and to invite them to identify their individual sources of exposure and to control them when possible. This is a concern for the managers in charge of large biomonitoring campaigns since there is a need of follow-up with the study participants whom's individual results may be considered as truly alarming (Haines et al., 2011).
In the absence of appropriate individual overexposure guidance value for SUIAS, it is still possible to use statistical criteria to identify situations of overexposure (e.g. Population's 95th percentile value) (Angerer et al., 2011; Ewers et al., 1999; Garnier et al., 2021; Vogel et al., 2019). However, for a naturally occurring contaminant such as iAs, a double challenge needs to be addressed when applying this approach: on the one hand, not to trivialize biomarker levels that are considered “normal” from a statistical point of view, but which are still associated with an elevated exposure and, likely, corresponding health risk; and on the other hand, not to stigmatize behaviors based on traditional diets such as high rice or seafood consumption, the latter associated to undeniable nutritionnal benefits(Mozaffarian and Rimm, 2006).
Indeed, interpretation of SUIAS values is still complicated by the fact that, for a large proportion of individuals, iAs intakes come mainly from food sources whose iAs levels are more or less controlled depending on the jurisdiction, or even strongly associated with diets that are perceived as being particularly healthy (e.g. gluten free diets) (Bulka et al., 2017). Moreover, certain kinds of seafood such as finfish, algae and a few shellfish can contain DMA as well as arsenosugars that are partially metabolized to DMA (CDC, 2017; Navas-Acien et al., 2011; Taylor et al., 2017). The occasional consumption of these seafoods can therefore cause a punctual increase in urinary concentrations of this metabolite. This increase could in turn result into overestimating the true iAs exposure and trigger an unjustified shift in dietary habits (National Academy of Sciences, 1999).
Under these circumstances, SUIAS results from populations primarily exposed to iAs through consumption of contaminated water have the advantage of being more representative of its toxic potential. In addition, these biomonitoring results can be interpreted in relation to the exposure associated to WHO's provisional guideline value (PGV) for iAs of 10 μg/L (World Health Organization, 2003), which is applied as a standard in several jurisdictions (Canada, USA, European Union, among others) (Council Directive (EU), 1998; Health Canada, 2006; U.S. EPA, 2001). This PGV is set on the basis of the best possible compromise between the cancer risk and the lowest achievable concentration in water, taking current DW treatment techniques into account (World Health Organization, 2017).
The objective of this study was therefore to propose an approach for the determination of a guidance value for biomonitoring data of SUIAS that allows the identification of individual study participants that may likely have experienced an iAs overexposure for which action of reduction can reasonably be envisaged at the individual scale. More specifically, the study aimed to 1) identify a range of SUIAS concentrations which could indicate a possible iAs environmental overexposure for North American and European populations; 2) assess the sensitivity and specificity of the different values within this range in order to orient decision-makers on the a priori selection of a final guidance value that allows an appropriate identification of overexposed individuals.
Section snippets
Framework
On the basis of the underlying premise that the WHO PGV in DW of 10 μg As/L constitutes a guideline that reflects an iAs exposure that is consensually considered as a level of which the exceedance should be avoided, the approach followed consists into predicting the range of SUIAS concentrations which would reflect 1) the consumption DW contaminated to or above the level of the PGV for As; and 2) the background exposure through the environmental, in particular, dietary sources of As in north
Selected studies
A total of 1137 studies were initially identified from the literature search strategy. Excluding duplicates, revising the titles and abstract for relevancy, adding relevant reference by the snowballing approach and applying the inclusion criteria (see Table 1) to the resulting group of references allowed to reduce this number to 14 studies, from which only 11 could be used for the upcoming SEM meta-analysis. Indeed, the remaining 3 had either missing summary statistics or had a high proportion
Originality
The objective of this study was to propose a range of potential SUIAS concentrations guidance values, reflecting a predetermined exposure level that should not be exceeded, that allows the individual identification of -and communication with- concerned biomonitoring study participants in North American and European populations, presumably sharing similar diets and environmental exposure. Rather than single guidance value, a range of values is proposed, more precisely 9–20 μg/L, with 15 μg/L
Conclusion
To the best of the authors’ knowledge, this is the first study to propose a range of guidance values that allows an exposure-based communication of human biomonitoring data on urinary As at the individual level. Concretely, such guidance value could be used during biomonitoring surveys to determine whether participants should be informed of their individual result and receive information and general advice to help determine, and if necessary control, their sources of iAs exposure. The
Credit author statements
Gabriela Ponce: Methodology, Formal analysis, Data curation, Visualization, Writing - original draft, writing – review & editing. Fabien Gagnon: Conceptualization, Methodology, Formal analysis, Writing - original draft, writing – review & editing. Marie-Hélène Bourgault: Methodology, Formal analysis, writing – review & editing. Michelle Gagné: Methodology, Formal analysis, Writing - original draft, writing – review & editing. Elhadji Anassour Laouan-Sidi: Formal analysis, Writing - original
Declaration of competing interest
The authors declare no conflict of interest.
Acknowledgements
Funding by Health Canada (MOA #4500370380) for the preliminary steps of this work is acknowledged. The authors would like to thank Dr Stéphane Buteau for his thoughtful comments throughout the course of this work. We thank Dr. David J. Thomas, Dr. Pierre Ayotte and Dr. Michèle Bouchard for providing the raw data of the studies of Calderon et al. (2013), Normandin et al. (2014) and Gagnon et al. (2016).
References (65)
- et al.
Human biomonitoring assessment values: approaches and data requirements
Int. J. Hyg Environ. Health
(2011) - et al.
Human biomonitoring: state of the art
Int. J. Hyg Environ. Health
(2007) - et al.
German Environmental Survey 1998 (GerES III): environmental pollutants in the urine of the German population
Int. J. Hyg Environ. Health
(2003) - et al.
Relationship of urinary arsenic to intake estimates and a biomarker of effect, bladder cell micronuclei
Mutat. Res. Rev. Mutat. Res. Arsenic: A Paradoxical Hum. Carcinog.
(1997) - et al.
Biological and behavioral factors modify biomarkers of arsenic exposure in a U.S. population
Environ. Res.
(2013) - et al.
Evaluation of human biomonitoring data in a health risk based context: an updated analysis of population level data from the Canadian Health Measures Survey
Int. J. Hyg Environ. Health
(2020) - et al.
Exposure of the French population to environmental chemicals ENNS 2006–2007 study
Toxicol. Anal. Clin.
(2017) - et al.
Dépistage, prise en charge et suivi des personnes potentiellement surexposées à l’arsenic inorganique, du fait de leur lieu de résidence–Recommandation de bonne pratique
Toxicol. Anal. Clin.
(2021) - et al.
Biomonitoring Equivalents for inorganic arsenic
Regul. Toxicol. Pharmacol.
(2010) - et al.
Biomonitoring equivalents: a screening approach for interpreting biomonitoring results from a public health risk perspective
Regul. Toxicol. Pharmacol.
(2007)