Using quantitative structure-activity relationship modeling to quantitatively predict the developmental toxicity of...

Area of Science:

• Environmental Toxicology
• Computational Chemistry
• Risk Assessment

Background:

• Developmental toxicity is crucial for human health risk assessment of chemical exposures.
• Animal testing for developmental toxicity is costly and time-consuming, leading to data gaps for many environmental contaminants.
• Halogenated compounds are prevalent environmental contaminants requiring toxicity profiling.

Purpose of the Study:

• To develop and validate a quantitative structure-activity relationship (QSAR) model for predicting developmental toxicity in halogenated compounds.
• To address data gaps in developmental toxicity profiles for risk assessment purposes.
• To provide a cost-effective alternative to traditional animal testing methods.

Main Methods:

• Utilized the OECD QSAR Toolbox (version 3.0) for chemical information curation.
• Developed a predictive model using linear regression with data from 35 curated halogenated compounds.
• Defined the model's applicability domain (AD) based on chemical category and structural similarity.
• Validated the model externally using seven halogenated chemicals not included in the training set.

Main Results:

• The developed QSAR model demonstrated strong predictive performance with R² of 0.79 and Q² of 0.77.
• External validation showed the model predicted lowest observed adverse effect level (LOAEL) values within a maximal threefold deviation for chemicals within the AD.
• The model successfully predicted developmental toxicity for compounds lacking experimental data.

Conclusions:

• QSAR modeling provides a reliable and efficient method to fill data gaps in developmental toxicity for halogenated compounds.
• This approach can significantly aid in the comprehensive human health risk assessment of environmental contaminants.
• The validated QSAR model is applicable for analyzing qualifying compounds where developmental toxicity information is incomplete.