Incorporating Bioavailability into Risk Assessment for Metal Mixtures: Results of a Comparative Evaluation
Eric Van Genderen (International Zinc Association), Emily Rogevich-Garman (Nickel Producers Environmental Research Association), Robert Dwyer (International Copper Association), Joseph Gorsuch (Copper Development Association)
While the concepts encompassing mixture toxicity and modeling have been around for decades, only recently have new approaches (“bioavailability models”) been expanded to consider metal mixture scenarios. Although current environmental regulations rarely require assessment of chemicals mixtures, the metals industries consider research on mixtures in the environment, and development of efficient and economic science-based risk assessment approaches for metals mixtures, essential to preparing for future regulatory demands and vital for ensuring adequate environmental protection. To this end, a comparative modeling evaluation was commissioned by the copper, nickel and zinc industries in an attempt to compare and contrast the available approaches for modeling metal mixtures in freshwater using the principles of bioavailability. The project culminated in a technical workshop (held 17-19 May 2012 in Brussels, Belgium), where results from the modeling approaches were presented to a broad group of scientists from the metals research community.
Four international research groups (US Geological Survey, Centre for Ecology and Hydrology, HDR|Hydroqual, and Japanese National Institute of Advanced Industrial Science and Technology), who are actively developing, testing and publishing a range of bioavailability-based approaches, were each provided the same mixture dataset (nearly 2,500 metal mixture and associated single metal exposures) to which they applied their respective modeling approaches. The modeling approaches represented the spectrum of complexity for single and multiple-site biotic ligand-type systems. Each modeling approach contained components for characterizing chemical speciation and organism sensitivity or toxicity. The overall dataset included chemical and toxicological information for mixture exposures from nine different studies, representing seven metals (Ag, As, Cd, Cu, Ni, Pb and Zn) in various binary, ternary and quaternary combinations using both spiked (laboratory and field waters) and ambient (field waters) experimental designs. For purposes of comparing the performance of the available approaches in the context of risk assessment, endpoints for individual treatments and samples were used instead of point estimates or hypothesis-testing endpoints.
The technical workshop was organized into three days of presentations and discussions. Each research group made presentations on their individual metals mixtures bioavailability models (e.g., background and development, methods applied to model the mixture dataset, in-depth discussion of the results of the model predictions and interesting findings and interpretations). Results of the comparative evaluation demonstrated that despite representing the spectrum of modeling complexity, the various approaches each provided consistent conclusions concerning interaction types and importance of accounting for bioavailability in metal mixture risk assessments. All of the modeling approaches were challenged by metal combinations where significant deviation from concentration addition hypotheses occurred (e.g., obvious antagonism observed in Cu-Cd and Zn-Cd mixtures). In addition, the biological variability associated with concurrent single-metal toxicity tests introduced notable uncertaintywhen interaction types and modeling performance were interpreted.
Research presentations were followed by group discussions on improving experimental design in metal mixture studies, commonalities and differences of the various approaches, cross-technology adaptation and future research needs. The discussions were successful in providing a platform for assessing the current state of the science for metal mixture bioavailability models, as well as prioritizing actions and next steps for future metal mixture research needs. Overall, approaches have been successfully developed to incorporate bioavailability into metal mixture risk assessment. In addition, this evaluation illustrated the importance of incorporating variability into effects predictions and the need to emphasize the science behind modeling biological complexity when communicating risk.
Scientific publications reviewing the approaches applied, results of the evaluation and key messages concerning the incorporation of bioavailability into risk assessments for metal mixtures will be submitted in the coming months. In addition, an abstract describing this evaluation has been submitted for the 2012 SETAC North America meeting (Special Symposium: 21st Century Risk Assessment).
Authors' contact information: firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org
Return to the Globe