Reliable predictive computational toxicology methods for mixture toxicity: toward the development of innovative integrated models for environmental risk assessment

Abstract

A main objective in the field of mixture toxicity is to determine how well combined effects are predictable based on the known effects of mixture constituents. Conducting toxicity tests for all conceivable combinations of chemicals, to understand all mechanisms in the combined toxicity of environmental pollutants, is virtually unfeasible due to cost- and time-consuming procedures. Therefore, predictive tools for mixture toxicity are required to be developed within the applicable range of predictive toxicology. The concept of concentration addition (CA) model is often considered a general method for estimating mixture toxicity at the regulatory level. In the long run, however, the possibility of toxicological synergism between mixture components actually occurs, especially from the no-effect level or non-toxic substances. This is ignored under the CA concept, and needs to be examined and integrated into existing addition models at a scientific level, this paper reviews existing integrated models for estimating the toxicity of complex mixtures in literature. Current approaches to assess mixture toxicity and the need for new research concepts to overcome challenges which recent studies have confronted are discussed, particularly those involved in computational approaches to predict mixture toxicity in an environment risk assessment based on mixture components.