Advancing Adverse Outcome Pathways for Integrated Toxicology and Regulatory Applications
Ksenia Groh, EAWAG and Knut Erik Tollefsen, NIVA
Author names and numbers cited refer to abstracts from the SETAC Basel meeting.
Adverse outcome pathways (AOPs) are conceptual frameworks for organizing biological and (eco)toxicological knowledge in a manner that supports the evaluation and extrapolation of biological perturbations reflecting the initiation (molecular initiating event) and early progression of toxicity across molecular, cellular and suborganismal levels (key events) towards an apical adverse outcome relevant to risk assessment and regulatory decision-making (Ankley et al. 2010). The AOP concept promotes a toxicological pathway-based vision relying on non-testing (computational) and testing (empirical) approaches to generate the data needed to evaluate chemical hazard, both for human health as well as ecotoxicological risk assessment.
The session “Advancing Adverse Outcome Pathways for Integrated Toxicology and Regulatory Applications,” held on 12 May at the SETAC Europe 24th Annual Meeting in Basel, Switzerland, focused on reviewing recent advances in AOP development as well as discussing the next steps to be undertaken in this growing research field. The talks and posters provided a broad overview of the current status and conceptual approaches for AOP development and application, along with diverse examples of specific activities and ongoing projects.
Several presentations covered the outcomes of a recent international workshop held in March 2014 in Somma Lombardo, Italy, which focused on exploring how to advance the use of AOPs for integrated toxicology and regulatory applications (for workshop overview, Vinas 29). The five workgroups at this workshop discussed the issues related to sequential stages in the AOP development and application process adapted within the program on AOP development launched by OECD and members of OECD Extended Advisory Group on Molecular Screening and Toxicogenomics (EAG MST) in 2012 (Aizawa 30).
The workgroups' themes were:
- Identification of research priorities for development of AOPs in ecotoxicology
- Strategic approaches and practical guidance for AOP development
- Weight-of-evidence evaluation of AOPs
- Promoting the acceptance and use of AOPs for regulatory applications
- Applying AOPs to support the development of informed approaches to testing and assessment.
During the session, the presentations on workshop results were complemented with contributions by other researchers working on related subjects.
A broader use of the AOP concept holds the potential to improve the understanding and prediction of chronic toxicity because development of respective AOPs would help to organize and appraise the existing knowledge. This would allow researchers to assess confidence in the predictive relationships, as well as to identify data gaps to guide further research. Elucidation of links between sublethal effects and population-relevant outcomes in the individuals would provide the basis for a broader and more meaningful inclusion of sublethal toxicity data into risk assessment frameworks. Importantly, improved mechanistic understanding would facilitate the development of alternative tests, as well as aid in extrapolation across species by promoting the reciprocal use of toxicity information generated in different species, and focus the testing on key targets associated with a particular AOP (Groh MO208).
The AOP framework is already being applied to advance the research on several topics of current concern in human and wildlife toxicology. As presented during the SETAC session, AOP-based strategies are used to (i) deepen our knowledge on the endocrine disrupting effects in species other than aquatic vertebrates and mammals, such as the invertebrate Daphnia magna, employing a combination of existing data with data generated by targeted experiments purposefully designed to fill the existing knowledge gaps (Song 34), (ii) explain differential sensitivity of several endangered sturgeon species to dioxin-like compounds by using AOP information to guide the selection of relevant molecular targets potentially responsible for species sensitivity differences (Doering 87), (iii) guide the research on developing alternative testing strategies for the fish early lifestage (FELS) test using zebrafish (Danio rerio) embryos to screen for important molecular initiating and key events such as thyroperoxidase or acetylcholinesterase inhibition, potentially leading to later adverse outcomes (Knapen 33) and (iv) define an alternative strategy for assessment of chemical potential to induce developmental neurotoxicity in humans on the basis of in vitro and in vivo assays using metabolomics analyses that focus on adverse outcomes relevant to neuronal development and disruption of several neurotransmitter pathways (Leonards 89).
Due to the large complexity of biological organization and toxicity responses across different species, identifying and describing AOPs represents a challenging task. Strategic approaches and best practice principles to guide AOP development have been formulated (Lettieri MO209). Several further contributions presented ways to get more sophisticated mechanistic information from in vitro test systems (Xia MO215), as well as methodology and novel computational approaches to study perturbation of gene expression that could aid in AOP development. Applications of diverse “-omics” techniques serve as an important route to elucidation of sequences of molecular events within an AOP. A large number of contributions demonstrated how these techniques can be used in AOP development, with examples mostly focused on elucidation of molecular mechanisms of action for diverse model toxicants. In this regard, it is important to note that AOPs themselves are conceptualized as being non-chemical-specific, however, elucidation of molecular responses to model toxicants may support the AOP development. The information collected through diverse chemical-focused case studies can later be combined and generalized into an AOP that denotes a plausible sequence of biological events that would lead to an adverse outcome following a certain perturbation of a particular molecular target.
Both traditional microarrays as well as high throughput sequencing can be used to investigate the mechanisms of toxicity to inform AOP development in model and non-model fish species (Santos 88, Basili MO211, Doering MO214). A combination of AOP-based vision, toxicogenomics and gene networks analysis demonstrated the importance of non-genomic signaling through membrane estrogen receptor as modulator of classical soluble ER-mediated biomarker responses (Denslow 91) and led to elucidation of putative AOPs triggered by ionic silver and silver nanoparticles (Vinas MO212) and brominated flame retardants (Vinas MO213) in fish, with potential implications for human health. An integrated computational biology approach based on network inference and utilization of information available across several levels of biological organization was suggested as a means to construct AOP-based predictive models for effects of single compounds as well as mixtures (Antczak 90). Finally, a web-based computational tool SeqAPASS (sequence alignment to predict across-species susceptibility) was presented, that allows rapid assessment of functional conservation of protein targets on the basis of sequence similarity, supporting species extrapolations beyond traditional model species (LaLone 86).
A wiki-based IT tool named AOP Wiki is currently under development by joint efforts of US Environmental Protection Agency, US Army Corps of Engineers, EU Joint Research Centre and other partners. In addition to its function as an open repository of AOP information, this resource is also expected to promote collective participation of a broader scientific and regulatory community in AOP development, evaluation, exploration and application, in particular through its open commenting functions but also through its ability to visualize interlinked systems of AOP networks. The AOP wiki holding an initial set of AOPs developed within the starting phase of OECD program on AOP development is expected to be released for public viewing in September 2014.
Once a particular AOP is described, the supporting weight-of-evidence and strength of predictive relationships between key events and adverse outcomes needs to be evaluated, which can be practically performed using modified Bradford-Hill criteria to assess the strength of experimental methods and ecological relevance of the observed responses, as has been demonstrated using atrazine as a case example (Solomon 32). Further, a simple numerical approach to visualize the uncertainty associated with different levels of pathway information collected in an AOP has been proposed as a means to facilitate AOP evaluation and "acceptance" by a regulatory community (Antczak 31). The range of regulatory applications possible for a particular AOP is defined by its completeness or maturation status. While incomplete AOPs can be used in first tier screening, such as formation of chemical categories, advanced quantitative AOPs with high level of certainty can be applied in full risk assessment. Thus, AOPs provide a foundation for the design of informed approaches to testing and assessment that can strategically deploy screening level analyses to effectively focus testing resources and progressively employ more resource-intensive assays aimed at reducing the uncertainty as required by risk assessment (Tollefsen MO210).
Although the number of contributions submitted to this session was relatively modest, the outstanding rate of attendance by conference participants and vivid discussions held in the session room demonstrated a high level of interest within the community of environmental scientists and regulators towards a systematic development and broader use of AOP framework in environmental toxicology research and risk assessment. Reflecting the growing interest in this active research field, a session will be held at the SETAC North America 35th Annual Meeting in Vancouver, Canada, and organization of further discussion forums at upcoming conferences is foreseen.
Author contact information: firstname.lastname@example.org; email@example.com
Return to the Globe