Statistical Learning in Preclinical Drug Proarrhythmic Assessment

TL;DR

This paper develops and evaluates statistical learning models, including ordinal logistic regression and random forest, to predict drug-induced Torsades de Pointes risk from preclinical data, enhancing drug safety assessment.

Contribution

It introduces a comprehensive statistical learning framework for predicting TdP risk, combining multiple models and validation techniques for improved accuracy and interpretability.

Findings

Models accurately predict TdP risk levels.

Outlier drugs identified align with literature.

Framework is suitable for drug safety assessment.

Abstract

Torsades de pointes (TdP) is an irregular heart rhythm characterized by faster beat rates and potentially could lead to sudden cardiac death. Much effort has been invested in understanding the drug-induced TdP in preclinical studies. However, a comprehensive statistical learning framework that can accurately predict the drug-induced TdP risk from preclinical data is still lacking. We proposed ordinal logistic regression and ordinal random forest models to predict low-, intermediate-, and high-risk drugs based on datasets generated from two experimental protocols. Leave-one-drug-out cross-validation, stratified bootstrap, and permutation predictor importance were applied to estimate and interpret the model performance under uncertainty. The potential outlier drugs identified by our models are consistent with their descriptions in the literature. Our method is accurate, interpretable, and thus useable as supplemental evidence in the drug safety assessment.