Virtual clinical QT exposure-response studies - A translational computational approach.

Area of Science:

• Cardiovascular pharmacology
• Computational biology
• Drug safety assessment

Background:

• Proarrhythmic risk assessment is shifting towards integrating clinical, non-clinical, and computational evidence.
• Current computational methods predict event incidence, but predicting concentration-response relationships for clinical endpoints like QTc is less explored.

Purpose of the Study:

• To predict the concentration-response relationship of QT interval changes (QTc) using computational models.
• To compare computational predictions with clinical trial data for drug-induced QTc prolongation.

Main Methods:

• Developed full-heart computational models simulating human cardiac populations.
• Predicted QT interval changes across a range of drug concentrations.
• Compared computational concentration-QTc relationships with clinical data for moxifloxacin, dofetilide, verapamil, and ondansetron.

Main Results:

• Computational models accurately predicted concentration-response slopes for three of four drugs (dofetilide, moxifloxacin, verapamil).
• Predicted critical concentrations for 10 ms QTc prolongation were comparable to clinical trial values (0.5-1x ratio).
• This in silico method demonstrated an ability to approximate human critical concentration values, outperforming other existing methodologies.

Conclusions:

• Computational modeling of virtual cardiac populations can generate data comparable to clinical findings.
• This approach can enhance pre-clinical and clinical drug safety evaluations.
• The models offer broad exposure ranges for improved trial design and pre-clinical to clinical translation understanding.