Environmental pollution is an issue that will be with us for generations to come unless we act now. The Lancet Commission estimates that pollution is currently the cause of 9 million premature deaths per year (16% of all deaths) disproportionally impacting the poor and vulnerable populations such as children and it significantly contributes to chronic non-communicable disease1. As a global, science-led pharmaceutical company, AstraZeneca is committed to working to predict and manage the environmental hazard and risks of pharmaceuticals that reach the environment2.
Pharmaceuticals are found in environmental waters almost everywhere that patients access medicines, but they are typically detected in the nanogram to low microgram per litre range and these are usually below the concentrations required to affect wildlife, for example. However, given the biological potency of some pharmaceuticals, we believe that even these trace concentrations are not without risk. Chemicals found in the environment at these levels are referred to as micro-pollutants and here we are referring to the impact of pharmaceutical chemicals at a systems biology level on aquatic wildlife.
One major challenge we face in tackling this issue is the lack of environmental toxicology data for pharmaceuticals in the public domain. Society uses a few thousand pharmaceuticals in medicines but knowledge about their potential effect on wildlife is still quite limited. For example, in Europe the reason for this lack of data availability is two-fold (i) the majority of human medicines being used were authorized by the European Medicines Agency (EMA) before the requirement for a comprehensive environmental risk assessment (ERA) came into force in 2006; and (ii) only medicines that are authorized through a centralized procedure have their environmental data published in European Public Assessment Reports (EPARs), no data being available for medicines authorized through national, decentralized and mutual recognition procedures. The lack of environmental toxicity data in the public domain, together with increased reports of their environmental presence through monitoring studies makes it difficult for stakeholders to determine the environmental risks posed by many medicines.
To address these knowledge gaps AstraZeneca has joined a research consortium funded under the Innovative Medicines Initiative (IMI), that brought together 13 pharmaceutical companies with leading academic and regulatory scientists, small and medium-sized enterprises (SMEs) and research institutes to try and identify the best way to prioritise which legacy pharmaceuticals that lack environmental data should be of the greatest environmental concern in Europe3.
This is not an easy task; it isn’t simply those most commonly used, or those detected in the environment at the highest concentrations, that post the most risk. Neither is it necessarily those that are the most potent. Working with partners at University of Exeter and supported by others at the universities of Brunel, Chalmers, Gothenburg, and with the support of an environmental consultancy wca environment, the consortium set out to collate all the publicly available environmental data on human medicines. We did so with a few simple questions in mind about the susceptibility of wildlife to the specific action of pharmaceuticals:
- Which is typically the most sensitive environmental species - is it fish, invertebrates or algae?
- Do properties such as the ability of the medicine to accumulate in wildlife and food chains make a difference to toxicity?
- How conserved are human drug receptors or targets in wildlife?
- Are wildlife species that have the same drug-target orthologues more susceptible?
- Does how the medicine acts have an impact?
- Can we classify and rank groups of pharmaceuticals if we know only about one medicine in that group?
In addition to these questions, we also collected patient use data in each European country to determine how much of each medicine was entering rivers and what environmental risk those medicines posed.
Our report published in Environment International is the largest single publicly available data-set underpinning our understanding of the environmental risks of pharmaceuticals1.
Our analysis of these data was complex, but a critical step forward to better understanding the impact of pharmaceuticals at a systems biology level in aquatic wildlife. We collated data from 204 algal studies, 160 daphnia studies and 148 fish studies and examined the target conservation for 549 human drug targets across these species to look at their potential susceptibility.
Through this analysis we can provide evidence for some relatively simple conclusions that will help us prioritise those pharmaceuticals that are of most concern. We suggest that:
- The presence or absence of drug-target in wildlife is key to predicting susceptibility to low concentrations of medicines. Indeed, if the drug-target is only in fish, then it is likely to be the most sensitive species to set the appropriate levels of concern, meaning that testing in daphnia and algae may not be necessary.
- If the pharmaceutical is lipophilic (attracted to fat) it may bioaccumulate to higher concentrations in aquatic species and food chains and can reach critical concentrations inside those organisms, that might have an adverse effect. Pharmaceuticals with a high lipophilicity tend to have effects at lower water concentrations.
- Some classes of pharmaceuticals are more difficult to predict their risk; we therefore recommend that hormones and antineoplastic agents should be prioritised for testing; especially those that are already on the market without any data.
- Pharmaceuticals that are used only in a small patient population and have no lipophilicity or mode of action concerns are unlikely to present an environmental risk. Our analysis suggests that access to accurate patient use data together with the use of country-specific dilution factors would help to prioritise medicines for testing. If concentrations are predicted to be higher than 100 ng per litre then we recommend that these medicines should be prioritised for testing.
- There are often several different pharmaceuticals available with the same mode of action; we suggest that where no representatives of a mode of action class have already been tested those pharmaceuticals with the highest use and lipophilicity should be prioritised for testing.
- We also emphasise that given the limited number of data that exist, and the modes of action that lack coverage, the review of available environmental data and associated risks is iterative and should be updated frequently.
Although there are not that many pharmaceuticals with data, from the data that we have compiled for the first time ever we were able to establish that the environmental risks related to patient use for most human medicines is low even with worst-case exposure assessments. Where risks were observed they were for those medicines designed to act against reproductive drug targets e.g. contraceptives, hormone replacement therapy and some cancer medicines. Although this is reassuring, there is more work to be done as outlined in our recommendations, tackling those that are identified as being of most concern first.
1. Landrigan et al., 2018. The lancet commission on pollution and health. Lancet, 391:462-512. https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(17)32345-0/fulltext?elsca1=etoc
2. AstraZeneca. 2018. Pharmaceuticals in the environment position paper. https://www.astrazeneca.com/content/dam/az/PDF/2018/A2E303_Pharmaceutical%20in%20the%20environment_A4_Final_V4.pdf
3. http://i-pie.org/ Innovative Medicines Initiative Joint Undertaking under iPiE Grant Agreement n° 115735, resources of which are composed of financial contribution from the European Union’s Seventh Framework Programme (FP7/2007-2013) and EFPIA companies’ in-kind contribution.
4. Gunnarsson et al., 2019. Pharmacology beyond the patient – The environmental risks of human drugs. Environment International. 12; 320-332 https://www.sciencedirect.com/science/article/pii/S0160412019309493