By Alice Marshall, PhD researcher in Environmental Toxicology at University of York & contributor of PREMIER WP2.1; and Caroline Moermond, Sr Policy Advisor at the National Institute for Public Health and the Environment (RIVM), lead of PREMIER WP4 and contributor of WP3.

With the continuous development of new and existing pharmaceuticals, investigation into the environmental legacy of active pharmaceutical ingredients (APIs) is a rapidly growing area of research. Following manufacture, testing and consumption, APIs are released into aquatic and terrestrial environments, with cocktails of API residues now routinely detected in surface waters, soils and sediments across the world. Ecotoxicity testing of APIs is therefore needed to assess their risks to the environment. However, the existing dossier of data describing environmental risk for individual APIs is highly variable in quality and transparency. During a PREMIER Masterclass, testing guidelines and quality evaluation of data were discussed.

Before approval, APIs that are used above certain amounts are subject to environmental testing. However, data collection and reporting is often unstandardised and inconsistent. Similarly, APIs marketed before 2006 are rarely supported by an environmental dossier. PREMIER therefore aims to fill existing data gaps for 25 priority APIs. For the existing data (mostly peer reviewed literature), reliability and relevance need to be assessed.

In view of these inconsistencies in datasets to date, work within PREMIER will perform new tests and evaluate existing data in an objective manner, using standardised methods wherever possible. The PREMIER project will therefore apply a suite of guidelines and evaluative tools to achieve this. For example, f or the testing of chemicals within regulatory frameworks,  OECD guidelines will be followed. Use of these guidelines aims to standardise laboratory testing techniques for environmental safety testing of chemicals to generate reliable, comparable, traceable, and reproducible data. They usually contain a very detailed description of the test design, including exposure scenarios, test species, test systems, and environmental conditions. OECD guidelines also contain a description of the data to be recorded, statistical analysis and how to report the results. Thus, they require highly transparent and detailed reporting of raw data to ensure full transparency. Alongside this, regulatory tests also need to be performed under Good Laboratory Practice (GLP), which is a quality system which provides guidelines on the organisational process, providing information on the planning, performance, monitoring, recording, archiving and reporting of results. OECD and GLP guidelines set a high standard for scientific protocol when performing environmental risk assessments, but also facilitate sharing of data between regulatory frameworks.

However, neither OECD guidelines or GLP standards ensure good quality science, correct study setup, correct interpretation of results or appropriate testing of non-standard endpoints which may be needed for specific modes of action. OECD/GLP studies are therefore not always reliable. For example,their design may not fit the substance (e.g., unsuitable exposure conditions, no solvent control), the conduct of the study may contain flaws (e.g., control mortality too high), errors may be made in interpretation or statistical analysis (e.g., solvent control statistics, outliers) or the validity criteria may not be met.

Thus, the quality or reliability of all data, including OECD GLP studies, should be assessed in a transparent and systematic way. Within PREMIER, the CRED method will be used for this. The CRED method contains 20 evaluation criteria (e.g., on test design, species, exposure conditions, statistics, availability of raw data) with guidance on how to apply these. The assessment results in a reliability score from 1 to 4, where a reliability of 1 means the study is reliable without restrictions, 2 means the study is reliable with restrictions, 3 indicates the study is not reliable, and 4 means there is not enough information to complete the reliability assessment. Studies with reliability 1 and 2 will be used for further assessment, in regulatory contexts and also within PREMIER.

With increasing pressure on natural resources and rapid pharmaceutical innovation, thorough environmental safety testing is an essential component in developing sustainable APIs. The PREMIER project involves prioritisation and screening of current and emerging APIs in various environmental systems with the ultimate aim of generating a good-quality, large, accessible and transparent environmental dataset, which, through use of the standardisation tools outlined here, can be used to help accurately evaluate potential environmental risk.