Gradient-boosting threshold identification for developmental toxicity: A BPAF case-study framework for early-stage...

Area of Science:

• Environmental Science
• Toxicology
• Risk Assessment

Background:

• Polymers rely on chemical additives, posing environmental and health risks.
• Bisphenol AF (BPAF), a Bisphenol A (BPA) substitute, is a growing contaminant of concern due to its endocrine-disrupting properties and developmental toxicity.
• Current risk assessments lack ecological relevance and early-stage accuracy.

Purpose of the Study:

• To develop and demonstrate an integrated, multi-domain assessment framework for early-stage risk evaluation of chemical contaminants.
• To apply this framework using Bisphenol AF (BPAF) as a case study.
• To derive a screening-level reference dose (RfD) and evaluate environmental exposure patterns for BPAF.

Main Methods:

• Combined machine learning-derived toxicological benchmarks with spatial analytics.
• Developed an integrated, multi-domain assessment framework.
• Derived a screening-level reference dose (RfD) and evaluated BPAF exposure patterns.

Main Results:

• Identified an imbalance in global BPAF monitoring data, with most data from Asia and Europe.
• Preliminary risk flags were identified in sediment and water samples from China and parts of Europe.
• Industrial and anthropogenic activities were identified as primary sources of BPAF contamination.

Conclusions:

• The developed framework effectively supports early-stage risk prioritization and monitoring of BPAF.
• Strengthened chemical oversight is needed, prioritizing under-reported regions and industrial watersheds.
• Predictive toxicology and spatial analysis are valuable tools for environmental risk assessment.