You are here:
Accepting the separation of toxicology and statistics in New Approach Methodologies in Ecological Risk Assessment
Citation:
Raimondo, S. Accepting the separation of toxicology and statistics in New Approach Methodologies in Ecological Risk Assessment . 12th World Congress on Alternatives and Animal Use in the Life Sciences, Niagara Falls, CANADA, August 27 - 31, 2023.
Impact/Purpose:
This presentation will contribute to a workshop on New Approach Methodologies (NAMs) for ecotoxicology as part of the 12th World Congress on Alternatives and Animal Testing in Niagara Falls, Canada on August 28th. The presentation highlights some challenges around the acceptance of NAMs during a focused workshop on regulatory acceptance and next generation education of alternatives to animal testing.
Description:
As we move away from animal testing, we move towards a future of models. In ecotoxicology, this requires a shift in the roles of toxicology and statistics; one that is underway yet faces resistance by many who find themselves uncertain on how to view the defensibility of models poised to replace familiar animal tests. The leading edge of toxicology has shifted towards cellular and molecular endpoints, which require models to translate to biological responses applicable for risk assessors. Statistical models based on probability and data trends become more important in the analysis phases of ERA to help quantify uncertainty around numeric threshold values. In a reality where no model is “right”, how do we agree when models are useful for us to succeed in reducing animal testing and protecting the environment? Using Interspecies Correlation Estimation (ICE) models for extrapolation of acute toxicity in aquatic organisms as an example, the pros of simplicity and quantifiable accuracy are weighed against the unknown, unquantifiable uncertainties of lost toxicological mechanisms. The validity of ICE models has been demonstrated by multiple, independent researchers, yet toxicologists struggle with the absence of toxicological mechanisms and model acceptance, creating impasse on moving the state of the science forward. The value and uncertainties of ICE models are weighed against the environmental consequences associated with limited data and inaction of government agencies. The challenges reviewed here highlight larger future challenges: As the ecotoxicology community remain at odds over simple statistical models, how will more complex NAMs ever be accepted?
