Key result
The simple linear model Bnet performed equally well or better than complex biophysical models in predicting torsadogenic risk, achieving a ROC AUC of 0.82 versus 0.67 and 0.60 for biophysical models in the 7-ion-channel dataset.
Why the study?
Does a simple linear model (Bnet) predict drug-induced Torsades de Pointes risk as well as complex biophysical cardiac models?
Does a simple linear model (Bnet) predict drug-induced Torsades de Pointes risk as well as complex biophysical cardiac models?
A simple linear model based on net ion-channel block predicts drug-induced Torsades de Pointes risk as well as or better than complex biophysical cardiac models, highlighting the value of simpler models in cardiac safety screening.
Simple models matched complex ones for ion-channel toxicity prediction; leaves open whether biophysical detail adds value beyond current datasets.
There is growing interest in applying detailed mathematical models of the heart for ion-channel related cardiac toxicity prediction. However, a debate as to whether such complex models are required exists. Here an assessment in the predictive performance between two established cardiac models, gold-standard and cardiac safety simulator , and a simple linear model B net was conducted. Three ion-channel data-sets were extracted from literature. Each compound was designated a cardiac risk category based on information within CredibleMeds. The predictive performance of each model within each data-set was assessed via a leave-one-out cross validation. In two of the data-sets B net performed equally as well as the leading cardiac model, cardiac safety simulator , both of these outperformed the gold-standard model. In the 3 rd data-set, which contained the most detailed ion-channel pharmacology, B net outperformed both cardiac models. These results highlight the importance of benchmarking models but also encourage the development of simple models.
No takes yet. Share an insight, caveat, or question.
Hitesh Mistry (2017) studied Drug-induced Torsades de Pointes (TdeP). Bnet simple linear model vs. Complex biophysical models (gold-standard and cardiac safety simulator) was evaluated on Area under the receiver operating characteristic curve (ROC AUC) for predicting TdeP risk. The simple linear model Bnet performed equally well or better than complex biophysical models in predicting torsadogenic risk, achieving a ROC AUC of 0.82 versus 0.67 and 0.60 for biophysical models in the 7-ion-channel dataset.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: