A probabilistic and context-dependent cell culture modelling framework for regulatory toxicology

Area of Science:

• Toxicology
• Cell Biology
• Regulatory Science

Background:

• Current in vitro models and New Approach Methodologies (NAMs) face limitations in accurately predicting human clinical outcomes.
• Oversimplified assumptions in existing systems neglect crucial physiological aspects like variability, interconnected pathways, and biological rhythms.

Purpose of the Study:

• To identify foundational physiological aspects often underrepresented in current in vitro models.
• To propose a novel probabilistic and context-dependent cell culture model for enhanced translational accuracy in regulatory toxicology.
• To reframe biological effects as probabilistic outcomes influenced by complex biological networks.

Main Methods:

• Identification of underrepresented physiological principles: degeneracy, interconnected pathways, individual variability, biological rhythms, and context dependency.
• Development of a probabilistic and context-dependent cell culture model integrating viability, functional fidelity, pathway mapping, temporal resolution, and Bayesian inference.
• Anchoring in vitro measurements to clinical benchmarks and incorporating patient perspectives.

Main Results:

• The proposed model accounts for physiological complexity, viewing biological effects as probabilistic outcomes.
• Integration of key physiological aspects enhances the translational relevance of in vitro toxicological assessments.
• The framework aims to bridge the gap between preclinical findings and clinical outcomes.

Conclusions:

• Embracing physiological complexity in NAMs is crucial for overcoming the translational bottleneck in toxicology.
• The proposed probabilistic model offers a pathway for developing more predictive, acceptable, and implementable NAMs.
• Advancing NAMs with these principles can lead to safer and more ethical toxicological assessments.