Taking adverse outcome pathways to the next level

Started more than 2 decades ago, the field of toxicology tends to transition from classical toxicology, focusing on the measurement of apical endpoints of toxicity in animal models, to predictive in vitro toxicology, relying on information on toxic mechanisms (NRC, 2007). This paradigm shift has been embodied, at least in part, in adverse outcome pathways (AOPs), first established in the area of ecotoxicology in 2010 (Ankley et al., 2010) and later on introduced in human toxicology as well. Basically, an AOP is a pragmatic tool to rationally and visually capture existing knowledge regarding the mechanistic basis of toxicity. Although conceptually very similar, the scope of an AOP is broader compared to the mode-of-action, as it can go up to the population level. Furthermore, while the mode-of-action tends to be chemical-specific and takes into account kinetic aspects, such as metabolism, AOPs are chemical-agnostic and describe a toxicological process from a purely dynamic perspective (Becker et al., 2015; Burden et al., 2015; Edwards et al., 2016; Leist et al., 2017; Perkins et al., 2015; Villeneuve et al., 2014a; Vinken et al., 2017). An AOP typically starts from a molecular initiating event (MIE) (i.e. a trigger of toxicity) and progresses through a series of key events (KEs), linked by key event relationships, ultimately resulting in a specific toxicological effect (Villeneuve et al., 2014a and 2014b). Different types of information can be used during AOP development, including in silico, in vitro, in vivo, clinical and epidemiological data (Delrue et al., 2016; Edwards et al., 2016; Vinken et al., 2017). AOP development ideally complies with guidance from the Organization for Economic Cooperation and Development (OECD) (OECD, 2016). In fact, the OECD, in collaboration with a number of other agencies, has established an electronic repository for AOPs, called the AOP Wiki. At present, the AOP Wiki contains more than 260 AOPs for a plethora of toxicological effects, both relevant for the fields of ecotoxicology and human toxicology (http://aopkb.org/). Many of these AOPs as well as additional ones have also been described in scientific literature. The number of AOP reports indeed has steadily increased in the last lustrum and is expected to further do so in the next few years.

Several potential applications for AOPs in the toxicology and risk assessment have been described. AOPs were initially intended to support regulatory decision-making based on the desire to make effective use of mechanistic data, particularly novel data streams that can be generated more rapidly and cost-effectively in a high-throughput format (Burden et al., 2015; Delrue et al., 2016; Edwards et al., 2016; Villeneuve et al., 2014a). AOPs can serve as the basis for generating integrated approaches to testing and assessment (IATAs) (Patlewicz et al., 2015; Tollefsen et al., 2014). While IATAs provide a platform for data integration and a means for targeted testing for a specific purpose, it is not necessarily framed by a mechanistic rationale. AOPs could be used to provide this mechanistic basis and thus to identify data gaps or to contextualize a diverse universe of existing data (Delrue et al, 2016, Tollefsen et al., 2014). AOPs may also facilitate chemical categorization by defining profilers that represent structural alerts associated with the induction of specific MIEs (Delrue et al., 2016). By doing so, AOPs can enable prioritization of chemicals for further testing and potency ranking (Burden et al., 2015; Patlewicz et al., 2015). This, in turn, could assist in the classification and labelling of chemicals (Patlewicz et al., 2015). Another appealing application includes the use of MIEs and KEs as the conceptual foundation for setting up batteries of in silico and in vitro tests that collectively allow accurate prediction of toxicity, with significantly higher sensitivity and specificity compared to stand-alone tests (Vinken et al., 2017).

The specific application of an AOP is de facto dictated by its degree of maturity. In this respect, the vast majority of currently available AOPs are descriptive (i.e. qualitative) in nature and focus on only 1 MIE and a linear series of adverse KEs. This may be sufficient for some purposes, yet most applications in toxicology and risk assessment require a more advanced level of AOP development. After all, toxicological effects are much more complex in reality than depicted in a simple AOP construct. Efforts in the upcoming years should be focused on further optimizing AOPs in order to make them fit for more general purposes in routine toxicology and risk assessment. An improvement recently came with the introduction of so-called AOP networks, which combine different AOPs that share at least 1 KE (Villeneuve et al., 2014a). However, currently available AOP networks, such as for liver steatosis toxicity (Mellor et al., 2016), disregard homeostatic adaptation, a process spontaneously activated by the organism to counteract inflicted adversity. Nevertheless, adaptation is equally important as adversity in determining the actual manifestation of toxicity. Risk assessors typically use numbers and thresholds rather than merely a descriptive series of mechanisms to evaluate safety of chemicals. Hence, quantification of AOPs should become a priority in the following years (Leist et al., 2017; Vinken et al., 2017). At present, quantification of AOPs seems mainly restricted to establishing dose/concentration-effect relationships, which as such strengthen confidence in proposed KEs, but that have only limited value for risk assessment per se. More specific quantitative parameters, mechanistically anchored in AOPs, should be applied or developed, such as the cholestatic index for evaluating bile acid-associated liver toxicity (Chatterjee et al., 2014). Furthermore, resulting quantitative AOP networks should be further embedded in structures that also consider other critical aspects of sound risk assessment, in particular kinetics and exposure elements. It can be anticipated that in-depth exploration and, in particular, optimization in the upcoming years will enable the development of the full potential of AOPs.

Abbreviations

AOP(s)

adverse outcome pathway(s)

IATAs

integrated approaches to testing and assessment

KE(s)

key event(s)

MIE(s)

molecular initiating event(s)

OECD

Organization for Economic Cooperation and Development