A Bayesian Method for Population-wide Cardiotoxicity Hazard and Risk Characterization Using an In Vitro Human Model

Area of Science:

• Toxicology and Pharmacology
• Stem Cell Biology
• Computational Modeling

Background:

• Human induced pluripotent stem cell (iPSC)-derived cardiomyocytes are valuable for chemical hazard testing.
• Interindividual variability in toxicodynamic sensitivity is known but not fully quantified population-wide.
• Previous models lacked quantitative characterization of population-wide toxicodynamic variability.

Purpose of the Study:

• To develop and apply a population-based iPSC-derived cardiomyocyte model combined with Bayesian analysis to quantify toxicodynamic variability.
• To assess cardiotoxic risk for pharmaceuticals and environmental chemicals across a human population.
• To enable more accurate in vitro-to-in vivo extrapolation for chemical safety assessments.

Main Methods:

• Utilized a population-based model with iPSC-derived cardiomyocytes from 43 humans.
• Tested 136 compounds (54 pharmaceuticals, 82 environmental chemicals).
• Employed hierarchical Bayesian population concentration-response modeling and toxicokinetic modeling for extrapolation.

Main Results:

• Quantified chemical- and phenotype-specific toxicodynamic variability factors, often exceeding the default value of 3 for functional phenotypes.
• Pharmaceuticals showed activity across all tested phenotypes.
• Over half of environmental chemicals exhibited activity, primarily positive chronotropy; margins of exposure generally exceeded 1000.

Conclusions:

• The developed in vitro-in silico model effectively characterizes population-wide toxicodynamic variability in cardiotoxic risk.
• Population variability-based safety margins for pharmaceuticals were narrower than expected, highlighting risks for sensitive subpopulations.
• Environmental chemicals showed low risk at current exposure levels, even for sensitive groups, based on derived margins of exposure.