Cumulative risk assessment: A European perspective on the state of the art and the necessary next steps forward

Authors


The first Europe-wide system for chemical assessment and management was introduced in 1967 with the Dangerous Substance Directive 67/548/EEC, which established a hazard-based classification and labeling system for chemicals on the European market. Subsequent risk-based regulatory frameworks comprised the 2 predecessors of the REACH Regulation (e.g., Directive 93/67/EEC and Regulation [EC] No 1488/94). The enactment of these and similar European-wide Directives and Regulations for chemical risk assessment, management, and mitigation resulted in a decrease of the concentrations of many classical toxicants found in European populations and ecosystems over the last decades, despite the continuous increase in chemical production and use. This is to be regarded as a huge success.

However, at the same time the sheer number of contaminants found in humans, nonhuman biota, and the various environmental compartments is continuously increasing, as is their spatial distribution: the typical exposure scenario is shifting from locally confined, high concentrations of individual compounds to a diffuse, widely spread blanket of contamination, composed of a myriad of individual pollutants each present at low concentrations.

Although approaches for the assessment, management, and mitigation of the impacts of local pollution from point sources are largely agreed on and widely applied on a routine basis, the assessment of diffuse pollution is still a major challenge for environmental policy and chemical management. To meet this challenge will require a move away from a narrow focus on individual pollutants, coarse acute individual or population level endpoints, the exclusive consideration of single emission sources, and exposure routes toward a broader, more holistic approach. This must be based on the analysis of combined effects of cumulative exposures to multiple stressors via all sources and exposure routes and has to consider sophisticated sublethal endpoints that inform on chronic toxicity.

The European discussion on combination toxicology and ecotoxicology has focused initially largely on the impacts of endocrine disrupter mixtures. However, a general characteristic of any pollutant cocktail is increasingly recognized: its total toxic effect exceeds that of each individual compound at the concentration at which it is present in the mixture. Consequently, the total effect of a mixture warrants general consideration to safeguard human health and the environment.

In the wake of these discussions toward the end of 2009, the European Council of Environmental Ministers published its conclusion on combination effects, requesting the European Union (EU) Commission to step up its chemical management and research efforts in the area. This resulted first in the state of the art report on mixture (eco)toxicology in 2010, followed by the mixture opinion of the European Scientific Committees in 2011 and finally the Communication of the European Commission back to the Council in 2012. In parallel, a whole array of reports and position articles by academia, industry (e.g., European Centre for Ecotoxicology and Toxicology), and transnational bodies (e.g., World Health Organization) were put forward, accompanied by a series of dedicated meetings, workshops, and conferences.

These articles and events show substantial common ground with respect to the scientific approaches for mixture toxicity assessment between human toxicology and ecotoxicology, as well as between the different stakeholders. It is unanimously agreed that for chemical mixtures, a classical tiered approach should be used and that the concept of concentration addition (CA) is a suitable approach for the first tier assessment, with the subsequent tiers using additional and/or more sophisticated models, e.g., independent action (IA), physiologically based pharmacokinetic (PBPK) models, interactions models, etc.

However, several scientific knowledge gaps remain to be closed in the coming years. This includes the consolidation and extension of empirical knowledge on the occurrence, toxicology, and ecotoxicology of environmentally realistic multicomponent mixtures of compounds that are composed of substances from different chemical groups and usage classes and do not share a similar mode or mechanism of action. As even the limited and scattered monitoring data available today indicate that the number of compounds present in an organism or at a specific site is often in the thousands, it will be critical to develop approaches on how to identify the drivers of (eco)toxicological impact: how can highly complex mixtures be simplified to facilitate appropriate risk management and mitigation measures?

Both CA and IA are based on the assumption that the compounds in a mixture do not interact, neither in their toxicokinetic nor in their toxicodynamic phase. Such interactions might lead to synergistic mixture toxicities that are substantially higher than predicted by CA, a phenomenon that is explicitly exploited during the development of chemical combination products (e.g., pesticides or pharmaceuticals). To ensure adequate protection of human health and the environment from unwanted mixture effects, it will hence be necessary to conduct a systematic “search for synergy”: under which conditions (number and types of compounds involved, [eco]toxicological endpoints triggered, species exposed) are more than concentration-additive mixture effects to be expected?

Predictive mixture hazard and risk assessment approaches are the only option for a proactive chemicals management. Such predictions rely not only on sound mixture-aware exposure assessment and modeling approaches, but also on the availability of high quality data on the (eco)toxicology of the individual substances that make up the pollution cocktail. Unfortunately, the current practice of castrating (eco)toxicological data sets to EC50 values and/or no observed effect concentration (NOEC) before publication is a major problem for mixture assessment. The use of existing (eco)toxicological knowledge on individual mixture components would be much improved if the complete concentration–response information—that formed the basis for generating EC50 and NOEC values in the first place—would be appropriately documented.

Most of the scientific challenges ahead are about known unknowns. They are to a large extent related to filling in gaps, i.e., the need to generate more and improved quantitative data on mixture exposure and (eco)toxicology, and to refine and extend the applicability domain of existing concepts. This is still a major undertaking but may be seen as a minor obstacle compared to the conceptual challenge posed by the issue of combination effects for the current European system of chemical regulation and management. This is currently structured into regulatory silos, along commercial use patterns: the European Food Standards Agency (EFSA) assesses pesticides, the European Chemicals Agency (ECHA) handles industrial chemicals and biocides, and the European Medicines Agency (EMA) deals with pharmaceuticals. And each of these European authorities interacts with the corresponding regulatory authorities of the individual 28 EU member states in its own, very specific way.

In stark contrast to this structure, the pollutant mixtures that humans and environmental organisms are exposed to in their everyday life do not follow the artificial boundaries defined by regulatory silos. They are composed of pesticides, industrial chemicals, biocides, and pharmaceuticals at the same time, not to mention nonintentionally produced chemicals and transformation products. Currently there is no European institution in place that could orchestrate the assessment of such exposure situations, with a mandate to set binding human health or environmental quality standards. The Water Framework Directive (WFD) could possibly play such a role at least for the aquatic environment, but it only provides an umbrella and a European WFD-authority does not exist.

These organizational silos of the European regulatory system also become obvious when taking a closer look at the political will to consider mixture effects in more recently enacted legislative frameworks. Although the new Biocide Regulation (EU) 528/2012 puts substantial broad emphasis on safeguarding human health as well as the environment from unwanted combination effects, the new Pesticide Regulation EC 1107/2009 limits mixture assessment to pesticide residues and their cumulative effect on human and animal health (Art 4), respectively to combination effects of active ingredients and coformulants (Art 29). The broader consideration of environmental mixture effects was thrown out during the final vote on the Regulation in the European parliament.

Holistic assessments of realistic exposures also require that the more technical inconsistencies between the different regulatory approaches are overcome. For example, in the EU the ecotoxicological effects of pesticides are assessed on the basis of their toxicity–exposure ratio (TER), which is calculated separately for each major organism group. In contrast, the closely related biocides are assessed on the basis of their ecosystem-wide PEC–PNEC ratio (Predicted Environmental Concentration vs Predicted No Effect Concentration). Although both approaches have the same conceptual basis—the comparison of an expected exposure with the ecotoxicity of a compound—the different implementations and nomenclatures render the assessment of mixtures comprising pesticides as well as biocides unnecessarily complicated or even impossible.

In view of the obvious need for giving the regulatory system a much-needed overhaul, it is encouraging that the communication of the EU Commission to the EU council envisages setting up an ad hoc working group of “relevant services and associated Agencies and Authorities (EFSA, ECHA, EMEA and EEA) to strengthen coordination across the different pieces of legislation and to promote the integrated assessment of priority mixtures, taking into account the risks of human and environmental exposure.” However, it remains to be seen whether such a group will indeed manage to develop an impact and improve the practice of chemical risk assessment and management in Europe.

The assessment and management of complex mixtures comprising a multitude of pollutants at low, individually perhaps nontoxic, concentrations runs the risk of apportioning blame, a situation in which each stakeholder solely focuses on a small subset of the compounds present, highlighting the comparatively small contribution of “his” or “her” compounds. This risk of a diffusion of responsibilities can only be avoided by creating a solid, consistent and agreed European-wide regulatory framework for an integrated assessment, management and mitigation of mixture risks, cutting across current regulatory silos, with well-defined roles and responsibilities of the different stakeholders and regulatory bodies.

  • Thomas Backhaus

  • Editor

  • Integrated Environmental Assessment and Management

  • University of Gothenburg, Sweden

  • Michael Faust

  • F&B Environmental Consulting, Bremen, Germany

  • Andreas Kortenkamp

  • Institute for the Environment, Brunel University, United Kingdom

Ancillary