Development of Drug-induced Arrhythmia Prediction System

Abstract

A number of drugs in clinical use and during development have had to be withdrawn because of their proarrhythmic side effects. These otherwise promising compounds are a risk for patients and force the pharmaceutical industry to bear enormous costs. For assessment of potential risk of drug-induced cardiac arrhythmia, the effects of a drug on the Human ether-a-go-go related gene (HERG) channel expressed in either mammalian cells or Xenopus oocytes, and the prolongation of action potential duration (APD) and QT interval in animals, are now routinely measured. However, there are enough examples to indicate that these tests are not accurate or sufficient to assess the proarrhythmic risk of drugs. Therefore, further development of prediction methods for risk assessment is awaited to increase safety in drug usage in clinics and in new drug development. An in silico strategy has particular merits for this purpose, because, different from purely experimental approaches, it can easily deal with the diverse actions of the drugs. The in silico assessment of a drug's proarrhythmic risk requires a method that provides quantitative and comprehensive comparison of the effects of different forms of I_Kr-blockade upon APDs and transmural dispersion of repolarization (TDR). Previously, we have developed various cardiac action potential models, as well as a novel risk-assessment method based on I_Kr-I_Ks, two-dimensional maps of APD and TDR. In the present paper, we introduce a new prediction system based on cardiac action potential simulation and model database. This system consists of registered models, accumulated calculation results, computer cluster and client software. Various procedures for simulation and analysis can be executed on the web. The developed prediction system may be a useful tool to predict the risk of drug-induced arrhythmia.