Key result
Adding demographic features to machine learning fails to improve cardiotoxicity prediction over structural data alone.
Why the study?
Predicting drug-induced cardiotoxicity remains challenging, and despite known differences across demographic groups, incorporating demographic factors into in silico prediction models remains difficult.
Incorporating demographic factors from spontaneous reporting data into in silico prediction models for drug-induced cardiotoxicity provides little additional predictive value over structural information alone.
Demographics from pharmacovigilance data add little to cardiotoxicity ML models; leaves open their value in prospective or richer datasets.
Predicting drug-induced cardiotoxicity remains one of the most important challenges in drug safety, contributing to a substantial share of clinical trial failures and postmarket withdrawals. While clinical evidence shows differences in adverse responses across sex, age, and body mass, incorporating demographic factors into in silico prediction models remains challenging. We developed CARBIDE (CARdiotoxicity Based on Integrated Demographic Evidence), a collection of 27 dataset variants derived from the FAERS pharmacovigilance database, to systematically evaluate whether meaningful structure-demographic interactions could be learned from spontaneous reporting data. Through systematic evaluation of different FAERS filtering criteria, cardiotoxicity definitions, and statistical methods, together with comprehensive ablation studies, we found that machine learning models failed to extract useful structure-demographic relationships. The models either learned population-level statistics or relied solely on structural information, with demographic features derived using our approach providing little additional predictive value. While these findings reveal fundamental limitations in using pharmacovigilance data for demographic-aware toxicity prediction, CARBIDE's systematic evaluation provides important insights for the field, helping guide future efforts toward more effective approaches in personalized cardiotoxicity prediction.
No takes yet. Share an insight, caveat, or question.
Iwan et al. (2026) studied Drug-induced cardiotoxicity. Demographic factors in in silico prediction models vs. Structural information alone was evaluated on Predictive value of structure-demographic relationships. Machine learning models incorporating demographic features from pharmacovigilance data provided little additional predictive value for drug-induced cardiotoxicity over structural information alone.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: