HPC Framework for Performing in Silico Trials Using a 3D Virtual Human Cardiac Population as Means to Assess...

Area of Science:

• Computational biology
• Cardiovascular research
• Pharmacology

Background:

• Traditional drug cardiotoxicity testing relies on animal models and lengthy clinical trials.
• Assessing pro-arrhythmic risk, particularly with drug combinations, remains a challenge.
• Existing in-silico methods often lack detailed anatomical and physiological representation.

Purpose of the Study:

• To develop and validate a high-performance computational framework for in-silico clinical trials.
• To assess the pro-arrhythmic risk of hydroxychloroquine and azithromycin, alone and in combination.
• To investigate drug-induced QT-prolongation and arrhythmia mechanisms in a virtual human heart model.

Main Methods:

• Utilized 3D biventricular human heart models with phenotypic and sex-specific variations.
• Performed electrophysiology simulations to analyze pseudo-ECGs, calcium dynamics, and activation patterns.
• Validated in-silico findings with in-vitro experiments on reanimated swine hearts using Visible Heart® methodology.

Main Results:

• The in-silico trials accurately predicted pro-arrhythmic risk (21.8% vs. 21% clinical risk).
• Identified transmural heterogeneity in action potential prolongation as a key mechanism for drug-induced arrhythmias.
• Demonstrated that common phenotype variants lead to distinct drug-induced arrhythmogenic outcomes.

Conclusions:

• The computational framework provides a rapid and reliable method for in-silico drug cardiotoxicity trials.
• This approach effectively reproduces complex cardiac electrophysiology in diverse virtual populations.
• The study highlights the potential to reduce animal use and accelerate clinical trial timelines for drug safety assessment.