Abstract

The use of class I antiarrhythmic drugs may be complicated by the presence of active metabolites. A simple technique to predict the clinical activity of these metabolites might help with the clinical use of these drugs. We tested the hypothesis that drug metabolites bind to the class I drug receptor, but that only clinically active metabolites bind appreciably at clinically observed concentrations. Using a radioligand assay, we determined whether 13 class I drug metabolites interacted with a receptor for class I drugs associated with the cardiac sodium channel. The radioligand was [3H]batrachotoxinin A20 benzoate. All 13 metabolites bound to the drug receptor with IC50 values of 2.7 to 375 microns, and a mean Hill number of 1.0 +/- 0.3. All of the seven active metabolites (N-acetylprocainamide, mono-N-dealkyldisopyramide, 5-hydroxypropafenone, N-desisopropylpropafenone, 0-demethylencainide, 3-methoxy-0-demethylencainide and desethylamiodarone) each bound to the receptor at concentrations approaching their clinical concentrations, whereas none of the six inactive metabolites (quinidine-N-oxide, 3-hydroxyquinidine, glycinexylidide, monoethylglycinexylidide, N-demethylencainide and N,0-demethylencainide) did. Using a relationship which correlates drug IC50 values and clinically observed drug concentrations we calculated predicted clinical concentrations if the metabolites were clinically active. A predicted/observed ratio < or = 10 correlates with 100% positive and negative accuracies whether a drug metabolite has clinical activity. Thus a simple radioligand assay predicts whether class I drugs have clinical activity.