Understanding Pro-Arrhythmic Effects of Drugs using Computational Models and Parameter Sensitivity Analysis

Abstract

Increased risk of ventricular arrhythmia is a dangerous side effect of many pharmacological agents. Often, drugs block the K+ channel responsible for rapid delayed rectifier current (IKr), leading to delayed of action potentials, prolongation of the QT interval, and increased arrhythmia risk. Some drugs, however, block IKr potently but are nonetheless safe. In addition, the effects of a drug on action potential morphology depend not just on the channel that is blocked, but also on the other channels present in the cell, a concept known as repolarization reserve. We have gained new insight into both phenomena through analysis of ventricular myocyte computational models with parameter randomization and multivariable regression. The most likely targets of a non-specific drug can be deduced from the relationship between action potential duration and drug concentration, if the data are compared to the parameter sensitivity analysis of an appropriate electrophysiological model. Simulations also provide insight into how the electrophysiological substrate of a ventricular myocyte affects the response of the cell to a drug that blocks IKr. Such a drug always prolongs action potential duration, but the effects can be either exacerbated or attenuated, depending on the characteristics of the other ion channels present. Specifically, simulations with a common human ventricular myocyte model suggest that the most important factors influencing the response to an IKr-blocking drug are: 1) the underlying density of IKr; 2) the density of slow delayed rectifier current IKs; 3) the voltage-dependence of IKr inactivation; 4) the density of L-type Ca2+ current, and 5) the kinetics of IKs activation. These simulations provide for a quantification of the important concept of reserve, and demonstrate how analysis of computational models can provide insight into the factors that influence adverse drug reactions.