Additional Considerations

It should be noted that the fundamental principles underlying an ERA of industrial chemicals have been primarily developed on the basis of an understanding of the environmental fate and behaviour of neutral compounds, which assumes generic exposure pathways and generic toxicological modes of action, such as non-polar narcotic, polar narcotic, and reactive. This is how ERA has been conducted in the EU for both ‘new’ and ‘existing’ substances (ECETOC, 2003a; Tarazona et al, 2009; Franco et al, 2010; Rayne and Forest, 2010). Consequently, the domain of applicability, with respect to the physical and chemical space defined for the tools, models, and regressions used to perform a traditional ERA, is thus likely to be limited, and not necessarily appropriate for chemical substances that are ionised at environmentally relevant pH. Indeed, Franco et al (2010) suggested that a significant fraction of industrial chemical substances that have been .pre-registered’ at the European Chemical Agency (ECHA) likely consist of ionisable organic chemicals. Nevertheless, it should be possible to apply the principles of a traditional ERA, whereby the .risk characterisation’ ratio (RCR), of the predicted environmental concentration (PEC) and the predicted no effect concentration (PNEC), being the < 1 is indicative of the absence of risk. Details on how to derive PECs and PNECs are given by the EC (1996, 2003) in its technical guidance document (TGD) and the books of van Leeuwen and Hermens (1995) and van Leeuwen and Vermeire (2007). Challenges arise, however, when estimating PECs and PNECs for chemicals that are ionised at environmentally relevant pH. The TGD, for instance, suggests that the water solubility of the ionised species can be orders of magnitude higher than the neutral form, and consequently recommends that input parameters, such as KOW, Henry's Law constant, and the sorption coefficients for soil, sediment and suspended solids need to be corrected in order to account for the fraction that is in the ionised form (EC, 1996, 2003). Unfortunately, the correction does not consider electronic interactions that may occur between the ionised species and environmental surfaces; therefore, the guidance provided in the TGD may be inappropriate, resulting in a possible overestimate of the PEC to be derived.

In a survey of more than 900 APIs listed in the Australian Medicines Handbook, the majority of APIs were found to be ionisable (64.2%), with the remainder comprising compounds that had a high molecular weight (14.9%) or were neutral (12.4%), always ionised (4.7%), miscellaneous (2.4%) or inorganic salts (1.3%) (Manallack, 2009). When mixtures, salts, and high molecular weight chemicals were removed from the list, 85% of small molecular weight (< 1,000 Da) APIs were estimated to be ionisable (Manallack, 2009). Consequently, improved methods for ERA of APIs, which are predominantly ionisable organic compounds, is an area of particular interest. Indeed, under Article 8(3)(g) of Directive 2001/83/EC relating to medicinal products for human use, there is a specific requirement to indicate the potential risks posed by a medicinal product for the environment; the evaluation must be provided as part of the application for marketing authorisation (EU, 2001). Consequently, in 2006, the European Medicines Agency (EMEA) published a guideline for industry regarding the data requirements and steps needed to perform an ERA of medicinal products targeted for human use (EMEA, 2006). The guideline describes a two-phase process for ERA of the API being marketed, based directly on patient use. The first phase (Phase I) is a screening exercise, aimed at singling out those APIs whose usage (i.e. environmental exposure) is not likely to be of concern, and therefore do not require specific evaluation. Generally, APIs having a maximum daily dose of > 2 mg/d will require assessment in a second phase (Phase II). That is an in-depth ERA, requiring laboratory data on the environmental fate (degradation, partitioning) and effects (chronic toxicity) of the API. Within Phase II there are two tiers, A and B: Tier A focuses on the most likely point of environmental emission and exposure, namely the fate and discharge of the API from a wastewater treatment plant with emphasis on the parent compound. Tier B addresses potential areas of concern highlighted from the results of Tier A, such as exposure of the terrestrial environment, high toxicity or bioaccumulation, and may include further analysis of the metabolites. Unfortunately, like the TGD, Tier A of the EMEA guideline does not appropriately describe how to estimate a PEC for ionising organic compounds such as most APIs.

Given that > 80% of all APIs are charged (Comer and Tam, 2007; Manallack, 2009) and the increased interest in ERA of ionisable organic compounds, it was decided to focus this report on chemicals used in the pharmaceutical industry. The scope was further restricted to those pharmaceuticals that are used for human health purposes and, therefore, are discharged down-the-drain into the environment (emission scenario). The task force believed that, by collecting and analysing a set of data on APIs used in human health only, the results might be of greater quantitative value. Nonetheless, many of the observations made in this report related to APIs for human use were expected to be generally applicable to all ionisable organic compounds that are more broadly used in commerce. While adopting this approach, the task force recommended that veterinary pharmaceuticals and ionisable organic agrochemicals (e.g. pesticides), which are discharged to the environment from diffuse sources, be addressed as a separate activity. As an incentive, the experience from the agrochemical industry has been captured in Chapter 5 of this report.

Additionally, based on the importance of the PEC estimate to trigger a Phase II assessment as part of the EMEA approach, the task force agreed to focus its efforts on investigating factors that might influence estimates of the PEC. These included various testing strategies aimed at quantifying the biodegradability of pharmaceuticals and their sorption behaviour. The combined understanding of biodegradation and sorption should help to understand removal by wastewater treatment plants. In the absence of empirical data, the applicability of QSARs was investigated in order to estimate various physical-chemical property data. Further, developments in multimedia fate modelling have led to improved prediction of the environmental fate and behaviour of ionisable organic compounds.

Finally, there is considerable interest in improving the understanding of bioaccumulation of ionisable organic compounds, with particular concerns being raised around the use of trigger values based on KOW as a metric for bioconcentration testing. At the time of preparing this report, however, there was a paucity of data that would enable this question to be adequately addressed. Consequently, while this is an important issue that needs to be resolved, it was beyond the scope of this task force. Nevertheless, current regulatory requirements have resulted in a significant number of bioconcentration tests to be conducted, and the new data so obtained may provide future opportunities for better assessing the mechanisms influencing the bioconcentration and bioaccumulation of ionisable organic compounds. It is likely that, as an improved understanding of the controlling factors that influence bioconcentration emerges, there will also be opportunities to revisit the relevance of KOW in this context.