¹Jaela Foster-Burns, ²Nan Miles Xi

¹Department of Biology, Loyola University Chicago, Chicago, IL 60660, USA

²Department of Mathematics and Statistics, Loyola University Chicago, Chicago, IL 60660, USA

Torsades de pointes (TdP) is an irregular heart rhythm as a side effect of drugs and may cause sudden cardiac death. A machine learning model that can accurately identify drug TdP risk is necessary. This study uses multinomial logistic regression models to predict three-class drug TdP risks based on datasets generated from rabbit ventricular wedge assay experiments. The training-test split and five-fold cross-validation provide unbiased measurements for prediction accuracy. We utilize bootstrap to construct a 95% confidence interval for prediction accuracy. The model interpretation is further demonstrated by permutation predictor importance. Our study offers an interpretable modeling method suitable for drug TdP risk prediction. Our method can be easily generalized to broader applications of drug side effect assessment.

torsades de pointes; drug safety assessment; machine learning; multinomial logistic regression

1) Roden, D. M. Drug-Induced Prolongation of the QT Interval. N. Engl. J. Med. 350, 1013–1022 (2004).

2) Xi, N. M., Hsu, Y.-Y., Dang, Q. & Huang, D. P. Statistical learning in preclinical drug proarrhythmic assessment. J. Biopharm. Stat. 32, 450–473 (2022).

3) Blinova, K. et al. International multisite study of human-induced pluripotent stem cell-derived cardiomyocytes for drug proarrhythmic potential assessment. Cell Rep. 24, 3582–3592 (2018).

4) Cavero, I. & Crumb, W. ICH S7B draft guideline on the non-clinical strategy for testing delayed cardiac repolarisation risk of drugs: a critical analysis. Expert Opin. Drug Saf. 4, 509–530 (2005).

5) Colatsky, T. et al. The Comprehensive in Vitro Proarrhythmia Assay (CiPA) initiative - Update on progress. J. Pharmacol. Toxicol. Methods 81, 15–20 (2016).

6) Liu, T. et al. Utility of Normalized TdP Score System in drug proarrhythmic potential assessment: A blinded in vitro study of CiPA drugs. Clin. Pharmacol. Ther. 109, 1606–1617 (2021).

7) Joshi, A., Dimino, T., Vohra, Y., Cui, C. & Yan, G.-X. Preclinical strategies to assess QT liability and torsadogenic potential of new drugs: the role of experimental models. J. Electrocardiol. 37 Suppl, 7–14 (2004).

8) James, G., Witten, D., Hastie, T. & Tibshirani, R. An introduction to statistical learning. (Springer, 2021).

9) The elements of statistical learning: Data mining, inference, and prediction, second edition. (Springer, 2009).

10) Core Team, R. & Others. R: a language and environment for statistical computing. R Foundation for statistical computing, Vienna (2013).

11) RStudio Team (2015) RStudio Integrated Development for R. https://www.scirp.org › reference › ReferencesPapershttps://www.scirp.org › reference › ReferencesPapers.

12) Efron, B. Bootstrap Methods: Another Look at the Jackknife. aos 7, 1–26 (1979).

13) Breiman, L. Random Forests. Mach. Learn. 45, 5–32 (2001).

14) Christophe, B. In silico study of transmural dispersion of repolarization in non-failing human ventricular myocytes: Contribution to cardiac safety pharmacology. Br. J. Pharm. Res. 7, 88–101 (2015).

Volume 05 Issue 10 October 2022

There is an Open Access article, distributed under the term of the Creative Commons Attribution – Non Commercial 4.0 International (CC BY-NC 4.0) (https://creativecommons.org/licenses/by-nc/4.0/), which permits remixing, adapting and building upon the work for non-commercial use, provided the original work is properly cited.