Pharmaceuticals in the environment

Effects on environmental species

The negative effect of pharmaceuticals on organisms in the environment is well known. One example is the feminisation of fish.

To be able to evaluate whether or not an ecosystem is at risk, not only an exposure assessment needs to be performed, but also an effects assessment. Usually, effects of substances on environmental organisms are tested in the laboratory. Several organisms from aquatic and terrestrial ecosystems may be selected to conduct tests on. This includes microorganisms, plants, invertebrates and vertebrate test organisms (e.g., fish). Testing of vertebrates is subject to strict regulations, aimed at refining, reducing and replacing (the 3Rs) the use of test animals. 

Traditionally, toxicity tests focus on whole organism endpoints, with survival, growth and reproduction being the most measured parameters. Fundamental to toxicity testing is the establishment of a concentration-response relationship, which relates the endpoint measured in the test organisms to exposure concentrations. 

Standardised test methods are described by the OECD. Regulatory bodies generally require that toxicity tests supporting the registration of new chemicals are performed according to internationally recognised guidelines. Ecotoxicity data may be retrieved from databases in regulatory frameworks or in peer reviewed literature. Tests in peer reviewed literature are generally not performed according to standardised test guidelines, but may be equally informative. The tests that are needed to perform an environmental risk assessment before authorising pharmaceuticals for the market are described in the European Medicines Agency guidelines (EMA, 2024). Tests on species from at least three trophic (food chain) levels are needed: algae, daphnia (water flea) and fish.

Whole organism testing

Laboratory tests are usually performed with one substance on one species at a time. This way, any interaction between substances or between species can be avoided and the results can be attributed unequivocally. Often, toxicity tests are performed during short-term experiments, focusing on acute endpoints like mortality. This is in sharp contrast with the field situation, where organisms are often exposed to relatively low levels of pharmaceuticals for their entire life span. Thus, experiments demonstrating the effects of chronic (long-term) exposure are more relevant. Relevant endpoints to characterise the effect on species could then be growth, reproduction, and other more ‘subtle’ endpoints. It should be realised though, that the terms 'acute' and 'chronic' have to be considered in relation to the length of the life cycle of the organism. An exposure duration of four days is acute for fish, but chronic for algae, comprising already of four generations.

The result of toxicity tests is a dose-response relationship, where the response in relation to an endpoint (e.g, mortality, number of offspring) is plotted against the exposure concentration. So-called concentration-response curves are then fitted, from which measures of the toxicity of the chemical can be calculated.

Alternatives for animal studies

To reduce the number of animal tests, in vitro bioassays are developed that make use of tissues, cells, or proteins (Hamers, 2019). In vitro bioassays are usually performed to screen individual substances or samples for specific bioactive properties. Besides this, computer models (in silico models) can also be used to predict toxicity. These models are based on a training set of experimental test data. The in silico models are not sophisticated enough yet to completely replace lab experiments but can be used to prioritise testing and avoid the development of certain classes of chemicals based on common toxicological traits.