It is one of the most debated questions in chemical risk assessment: can exposure to chemicals such as pesticides cause ill health in humans? Scientists currently rely on experimental toxicology evidence, such as animal or cell studies, to assess their safety. A new approach proposed by EFSA will help further our understanding by making better use of epidemiological studies on human health.
Human epidemiological studies have suggested an association between exposure to certain chemicals and human diseases. However, as an association does not necessarily prove a cause-and-effect link, firm conclusions cannot easily be drawn from epidemiological studies. This means that such studies are in many cases of limited use in determining whether a chemical can ultimately represent a risk factor for a disease.
In 2013 EFSA published the results of a literature review on epidemiological studies linking exposure to pesticides and health effects. Since then the Authority’s pesticide experts have been exploring how results from epidemiological studies can be integrated into pesticide risk assessments. Although this work is relevant for all chemicals, it is particularly important for EFSA’s assessments of applications for approval of pesticides, as EFSA is obliged by EU regulations to evaluate epidemiological findings.
As part of this effort, EFSA’s Panel on Plant Protection Products and their Residues has tested a method that could enable risk assessors to establish a biological cause-and-effect link between exposure to chemicals such as pesticides and ill health. Dr. Susanne Hougaard Bennekou, chair of the working group that developed the method, and Dr. Andrea Terron, an EFSA staff scientist specialising in pesticides, explain the significance of the work.
Dr. Susanne Hougaard Bennekou
What was the aim of the project?
Pesticide risk assessors want to make better use of the information contained in epidemiological studies. To do this, we need to find a way of confirming – or not confirming – the associations suggested by the studies. This task is particularly difficult when the data relate to complex human diseases. Sometimes there are so many factors involved that it is just not possible to confirm the chemical-disease associations suggested in epidemiological studies.
In our Scientific Opinion we have applied a framework that could establish whether there is a plausible link between the cause – a chemical coming into contact with and affecting an organism at cellular level – and a subsequent chain of events leading to the effect i.e. disease. In other words, to establish whether a specific sequence, or pathway, of events represents a human health hazard – and consequently identify chemicals that need to be considered as potential risk factors for the disease.
This conceptual framework is known as the adverse outcome pathway (AOP). It was developed by the scientific community and implemented through the OECD to improve understanding of how chemicals induce adverse effects, but this is the first time that it has been adapted specifically for improving the use of epidemiological data in risk assessment.
Dr. Andrea Terron
How did you apply the AOP?
We designed prototype AOPs for Parkinson’s disease and infant leukaemia, two diseases for which the 2013 report showed consistent associations with pesticide exposure. We were able to plot a sequence of events – and event relationships – using different chemicals including pesticides.
And what did you find?
It is clear that the framework is an effective tool for establishing an association between exposure to a chemical and a disease. The overall weight of evidence from the AOPs for both Parkinson’s disease and infant leukaemia showed a strong link between the initial interaction – known as the molecular initiating event or MIE – and the adverse outcome.
So have you established a link between exposure to chemicals and these two diseases?
No, we have not established a link between exposure to a chemical and these two diseases. The AOP conceptual framework allows us to assess the plausibility of an association with a single chemical, but this is a long way from showing that a chemical causes a disease or even that it represents a risk factor – that requires a full risk assessment. For example, a pesticide will not necessarily have an adverse effect when people are exposed to it in real conditions – it has to reach specific cells to have an effect and also at a dose high enough to trigger the adverse outcome. Even then, exposure could be just one risk factor among multiple others – environmental and/or genetic – for developing Parkinson’s disease and infant leukaemia.
But this is a step forward for the risk assessment of chemicals, including pesticides?
Yes, it is. Although we are only just starting to understand the applicability of the AOP framework as a tool in risk assessment, it gives us the opportunity to contextualise all available information. The AOP is not intended to be used as a substitute for the data required to assess a pesticide but rather to serve as a complementary tool. The AOP therefore represents a shift in the risk assessment paradigm towards an approach that incorporates human health outcomes in the hazard identification process. The AOP method could also help to identify data gaps for each step along the pathway and to guide future testing strategies for hazard identification and characterisation.
This interview summarises and simplifies the main elements of a highly complex piece of scientific work. For a full, in-depth understanding of the issues, please consult the scientific opinion adopted by EFSA’s Panel on Plant Protection Products and their Residues.
- Investigation into experimental toxicological properties of plant protection products having a potential link to Parkinson’s disease and childhood leukaemia
What is epidemiology?
Epidemiology concerns the analysis of the patterns, causes (e.g. smoking, alcohol, virus, exposure to specific chemicals) and effects of health and disease conditions in defined groups of people. Epidemiology also defines how often diseases occur in different groups of people and why. The information provided is used to prevent illness, and in the case of exposure to chemicals to update information on the hazardous properties of a chemical and on the actual risk under past or current exposure conditions.