You are here:
Using Chemical Structure Information to Develop Predictive Models for In Vitro Toxicokinetic Parameters to Inform High-throughput Risk-assessment
Citation:
Pradeep, P., G. Patlewicz, R. Pearce, J. Wambaugh, B. Wetmore, AND R. Judson. Using Chemical Structure Information to Develop Predictive Models for In Vitro Toxicokinetic Parameters to Inform High-throughput Risk-assessment. Computational Toxicology. Elsevier B.V., Amsterdam, Netherlands, 16:100136, (2020). https://doi.org/10.1016/j.comtox.2020.100136
Impact/Purpose:
The toxicokinetic (TK) parameters fraction of the chemical unbound to plasma proteins and metabolic clearance are critical for relating exposure and internal dose when building in vitro-based risk assessment models. However, experimental toxicokinetic studies have only been carried out on limited chemicals of environmental interest (~1000 chemicals with TK data relative to tens of thousands of chemicals of interest). Data gap filling techniques such as read-across and quantitative structure-activity relationships (QSARs) are commonly used to predict hazard in the absence of empirical data. This work evaluated the utility of chemical structure information to predict TK parameters in silico; development of cluster-based read-across and QSAR models of fraction unbound (regression) and intrinsic clearance (classification and regression) using a dataset of 1487 chemicals; utilization of predicted TK parameters to estimate uncertainty in steady-state plasma concentration (Css); and subsequent in vitro–in vivo extrapolation (IVIVE) analyses to derive bioactivity-exposure ratio (BER) plot to compare human oral equivalent doses (OEDs) and exposure predictions using androgen and estrogen receptor activity data for 233 chemicals as an example dataset. The results demonstrate that fraction unbound is structurally more predictable than intrinsic clearance. The model with the highest observed performance for fraction unbound in plasma had an external test set RMSE/σ = 0.62 and R2 = 0.61, for intrinsic clearance classification had an external test set accuracy = 65.9%, and for intrinsic clearance regression had an external test set RMSE/σ = 0.90 and R2 = 0.20. The models were benchmarked against the ADMET Predictor software. Finally, the BER analysis allowed identification of 38 out of 233 chemicals for further risk assessment demonstrating the utility of these models in aiding risk-based chemical prioritization.
Description:
The toxicokinetic (TK) parameters fraction of the chemical unbound to plasma proteins and metabolic clearance are critical for relating exposure and internal dose when building in vitro-based risk assessment models. However, experimental toxicokinetic studies have only been carried out on limited chemicals of environmental interest (~1000 chemicals with TK data relative to tens of thousands of chemicals of interest). Data gap filling techniques such as read-across and quantitative structure-activity relationships (QSARs) are commonly used to predict hazard in the absence of empirical data. This work evaluated the utility of chemical structure information to predict TK parameters in silico; development of cluster-based read-across and QSAR models of fraction unbound (regression) and intrinsic clearance (classification and regression) using a dataset of 1487 chemicals; utilization of predicted TK parameters to estimate uncertainty in steady-state plasma concentration (Css); and subsequent in vitro–in vivo extrapolation (IVIVE) analyses to derive bioactivity-exposure ratio (BER) plot to compare human oral equivalent doses (OEDs) and exposure predictions using androgen and estrogen receptor activity data for 233 chemicals as an example dataset. The results demonstrate that fraction unbound is structurally more predictable than intrinsic clearance. The model with the highest observed performance for fraction unbound in plasma had an external test set RMSE/σ = 0.62 and R2 = 0.61, for intrinsic clearance classification had an external test set accuracy = 65.9%, and for intrinsic clearance regression had an external test set RMSE/σ = 0.90 and R2 = 0.20. The models were benchmarked against the ADMET Predictor software. Finally, the BER analysis allowed identification of 38 out of 233 chemicals for further risk assessment demonstrating the utility of these models in aiding risk-based chemical prioritization.
URLs/Downloads:
DOI: Using Chemical Structure Information to Develop Predictive Models for In Vitro Toxicokinetic Parameters to Inform High-throughput Risk-assessment