The use of benchmark dose uncertainty measurements for robust comparative potency analyses

Area of Science:

• Toxicology
• Computational Biology
• Genotoxicity Testing

Background:

• The Benchmark Dose (BMD) method is standard for dose-response analysis, using upper (BMDU) and lower (BMDL) confidence bounds for uncertainty.
• Accurate interpretation of BMD uncertainty for comparative potency analysis across different conditions is underexplored.

Purpose of the Study:

• To re-evaluate a genotoxicity dataset comparing chemical potency with and without S9 metabolic activation.
• To demonstrate the limitations of using BMD point estimates for comparative potency analysis.
• To introduce a statistically robust method using BMDL and BMDU for comparative potency.

Main Methods:

• Reanalyzed in vitro genotoxicity data for TK6 cells exposed to 10 clastogens +/- S9.
• Employed BMD modeling with PROAST v69.1 software.
• Calculated "S9 potency ratio confidence intervals" using BMDL and BMDU metrics.
• Utilized unsupervised hierarchical clustering to group compounds based on S9 effects.

Main Results:

• BMD point estimates can misrepresent potency differences between test conditions.
• "S9 potency ratio confidence intervals" offer a more statistically robust comparison of relative potency.
• Hierarchical clustering identified four distinct S9-dependent response patterns: potentiation, no effect, and diminution.
• The study highlights the critical role of BMD uncertainty measurements in comparative analyses.

Conclusions:

• BMD uncertainty measurements are essential for accurate comparative potency analysis between test conditions.
• A stepwise approach for BMD modeling in comparative potency studies is proposed.
• This methodology improves the statistical rigor of interpreting S9 metabolic activation effects in genotoxicity assessments.