Identification of Real-Life Mixtures Using Human Biomonitoring Data: A Proof of Concept Study

Area of Science:

• Environmental Health
• Toxicology
• Biostatistics

Background:

• Assessing human health risks from chemical mixtures is challenging due to vast exposure combinations.
• Human biomonitoring (HBM) provides data on internal chemical body burdens.
• Network analysis offers a method to visualize and understand complex exposure patterns.

Purpose of the Study:

• To explore the utility of network analysis for human health risk assessment using HBM data.
• To identify groups of correlated biomarkers representing real-life chemical mixtures.
• To evaluate the added value of network analysis for exposure and regulatory risk assessment.

Main Methods:

• Applied network analysis to HBM datasets from four European countries (Belgium, Czech Republic, Germany, Spain).
• Analyzed datasets with varying study populations, designs, and chemical analytes.
• Conducted sensitivity analysis to assess the impact of creatinine standardization methods.

Main Results:

• Network analysis successfully identified densely correlated groups of biomarkers (communities) within HBM data.
• The approach revealed consistent patterns across datasets of diverse origins.
• Demonstrated the ability to visualize and understand co-exposure to chemical mixtures.

Conclusions:

• Network analysis is a valuable tool for interpreting HBM data in the context of chemical mixtures.
• Identified biomarker communities inform regulatory risk assessment and the design of mixture exposure studies.
• This method enhances our understanding of real-world chemical exposures and their potential health implications.