Torsades de pointes liability inter-model comparisons: the experience of the QT PRODACT initiative.

Safety pharmacologists from the Japanese pharmaceutical industries and contract laboratories made a database to evaluate drug effects on the QT interval in 2005. This QT PRODACT project was a prospective study of 12 QT-prolonging (positive) drugs and 10 non-prolonging (negative) drugs to evaluate the specificity and sensitivity of several in vivo and in vitro animal models: in vitro guinea pig papillary muscle action potential recordings and in vivo ECG recordings in unanesthetized or anesthetized beagle dogs, cynomolgus monkeys and miniature pigs. In guinea pig papillary muscle action potential recordings, positive drugs showed lengthening of the action potential duration (APD). By using a new measure to detect triangulation of the action potential configuration, an IKr blocking activity of drugs with Ca channel blocking action was detected. All in vivo studies showed a QT-prolonging effect of greater than 10% for the positive drugs. These in vivo models were useful to distinguish positive from negative drugs. The QT PRODACT project showed reliability and sensitivity of the experiments to detect positive drugs. The proarrhythmic effects of these positive drugs could not be detected even though, in some animal models (e.g., unanesthetized monkey), torsades de pointes (TdP)-type arrhythmias were shown by terfenadine. We compared in vivo arrhythmia models for proarrhythmia. The halothane-anesthetized open chest coronary occlusion-reperfusion canine model, the halothane-adrenaline arrhythmia model and the chronic AV block dog models seemed to be useful to detect the arrhythmogenic potential of QT-prolonging drugs.

Comparison of four rate-correction algorithms for the ventricular repolarization period in assessing net effects of IKr blockers in dogs.

The utility of corrected and uncorrected QT interval changes for assessing net repolarization delay by I(Kr) (a rapid component of delayed rectifier K(+) currents) blockers was assessed in halothane-anesthetized dogs using the electrocardiogram and monophasic action potential (MAP) recordings with electrical ventricular pacing. Intravenous administration of dl-sotalol (0.2 - 2 mg/kg) prolonged the MAP duration and RR interval, while terfenadine (3 mg/kg) increased the MAP duration but transiently shortened RR interval. The order of correlation coefficient between the MAP duration at a pacing cycle length of 400 ms and MAP duration itself or that with arithmetical correction was uncorrected > Van de Water = Matsunaga > Fridericia > Bazett. These results suggest that Matsunaga's and Van de Water's formulae would better predict the net repolarization delay in the in vivo canine model. Also, the risk of drug candidates that may prolong the QT interval should be judged by change in uncorrected QT interval as well as corrected QT interval.

Acute hypokalemia may not be an effective way to sensitize the in situ canine heart for sparfloxacin-induced long QT syndrome.

Extents of the sparfloxacin (3 - 10 mg/kg, i.v.)-induced QT interval prolongation under normokalemic and hypokalemic conditions were assessed in halothane-anesthetized beagle dogs (n = 5). The hypokalemic condition was induced by an oral administration of furosemide (200 mg/kg per day) for 3 days, which decreased the serum potassium concentration from 3.65 +/- 0.13 to 2.35 +/- 0.13 mM (P < 0.05). However, the decrease of potassium concentration by itself did not affect the extent of the sparfloxacin-induced QT interval prolongation. These results indicate that acute hypokalemia may not severely sensitize the in situ heart for drug-induced long QT syndrome as previously thought.

Halothane sensitizes the guinea-pig heart to pharmacological IKr blockade: comparison with urethane anesthesia.

Potential utility of halothane-anesthetized guinea pigs for detecting drug-induced repolarization delay was analyzed in comparison with urethane-anesthesia (n = 4 for both groups). Basal QT interval was significantly greater under halothane-anesthesia than urethane-anesthesia (192 +/- 7 vs 132 +/- 5 ms, respectively), whereas the reverse was true for the heart rate (190 +/- 7 vs 248 +/- 11 beats/min, respectively). The typical I(Kr)-blocker dl-sotalol (0.1 to 3 mg/kg, i.v.) induced dose-related bradycardia and QT interval prolongation under each anesthesia. The extent of maximum prolongation in the QT interval was greater under halothane-anesthesia than urethane-anesthesia (+101 +/- 15 vs +49 +/- 3 ms, respectively), whereas that of peak change in the heart rate was smaller under the former than the latter (-49 +/- 8 vs -63 +/- 5 beats/min, respectively). Pretreatment of the animals under urethane-anesthesia with the selective I(Ks) blocker chromanol 293B (n = 6) increased the extent of the dl-sotalol-induced QT interval prolongation to +57 +/- 8 ms, which was only 0.56 times of that under the halothane-anesthesia, whereas the pretreatment increased the peak change in the heart rate to -76 +/- 12 ms. These results indicate that the halothane-anesthesia may effectively sensitize the guinea-pig heart to pharmacological I(Kr) blockade.

Reduction of repolarization reserve by halothane anaesthesia sensitizes the guinea-pig heart for drug-induced QT interval prolongation.

The utility of halothane-anaesthetized guinea-pigs as an in vivo model for predicting the clinical potential of a drug to induce QT interval prolongation was assessed using the electrocardiogram and monophasic action potential (MAP) recordings with electrical ventricular pacing. Intravenous administration of D-sotalol (0.3 mg kg(-1)) and terfenadine (0.3 mg kg(-1)), blockers of a rapid component of delayed rectifier potassium currents, prolonged the QT interval by 32+/-7 and 23+/-6 ms, respectively, whereas chromanol 293B (1 mg kg(-1)), a blocker of a slow component of delayed rectifier potassium currents, lengthened it by 33+/-8 ms. The extent of the QT interval prolongation by these drugs was greater than those in previous reports using pentobarbital-anaesthetized guinea-pigs. The MAP duration at the control was shortened by decreasing the pacing cycle length from 400 to 200 ms, but the MAP duration at each cycle length was prolonged by D-sotalol. The formulas of Van de Water, Matsunaga, Fridericia and Bazett showed good correlation of the repolarization period when compared with the MAP duration at a pacing cycle length of 400 ms. The halothane-anaesthetized guinea-pig model may possess enough sensitivity to detect drug-induced QT interval prolongation, indicating that halothane anaesthesia can reduce the repolarization reserve of the heart in vivo.

The antipsychotic and antiemetic drug prochlorperazine delays the ventricular repolarization of the in situ canine heart.

Electropharmacological effect of the antipsychotic and antiemetic drug prochlorperazine was assessed using the halothane-anesthetized in vivo canine model (n = 5). Up to 10 times higher than the clinically relevant doses of prochlorperazine (< or = 3 mg/kg, i.v.) did not induce cardiohemodynamic collapse in the model. Meanwhile, clinically relevant to supratherapeutic doses (0.3 - 3 mg/kg, i.v.) prolonged the ventricular repolarization period in a dose-related and reverse-use dependent manner that could become proarrhythmic substrates. Thus, caution has to be paid on the use of prochlorperazine particularly for patients with risks of the elevated plasma drug concentration, compromised cardiac repolarization, and/or frequent ventricular premature beats.

Electropharmacological effects of a spironolactone derivative, potassium canrenoate, assessed in the halothane-anesthetized canine model.

While aldosterone receptor blockers improve survival of patients with congestive heart failure, spironolactone and its derivatives were recently shown to block ether-a-go-go-related gene (HERG) channels and native IKs and IKr currents in guinea pig ventricular myocytes. In this study, we examined in vivo electropharmacological effects of an active derivative of spironolactone, potassium canrenoate, using a halothane-anesthetized canine model. Potassium canrenoate was intravenously administered in three doses of 1, 10, and 100 mg/kg per 10 min with a pause of 20 min between doses (n = 5). The low dose hardly affected any of the cardiovascular parameters. The middle dose, a clinically recommended daily maximum i.v. dose, slightly inhibited the intraventricular conduction. The high dose decreased the heart rate, ventricular contraction and blood pressure, delayed the atrioventricular and intraventricular conduction, and prolonged the ventricular repolarization and refractory period. Increment in the refractoriness by the high dose was greater than that in the repolarization, resulting in the reduction of ventricular electrical vulnerability. This unique electrophysiological profile of potassium canrenoate may in part contribute to the favorable clinical results, whereas caution has to be paid on the cardiohemodynamic actions, particularly for patients with risk of elevated plasma drug concentration.

Effects of class I antiarrhythmic drugs on the digitalis-induced triggered activity arrhythmia model: a rationale for the short-term use of class I drugs against triggered arrhythmias.

Delayed afterdepolarization (DAD)-induced triggered activity has been considered to be one of the generation mechanisms of ventricular arrhythmias in the presence of intracellular Ca2+ overload. In this study, we analyzed the antiarrhythmic effects of class I antiarrhythmic drugs, namely, disopyramide, procainamide, mexiletine, and flecainide, on a recently developed DAD-induced triggered arrhythmia model, which consists of a canine ventricular septum preparation cross-circulated with a blood-donor dog. After intravenous administration of ouabain to the donor dog, triggered arrhythmias were consistently induced in the cross-circulated preparation by train stimulation (cycle length: 300 ms; train number: 15). Intracoronary administration of disopyramide, procainamide, mexiletine, and flecainide as well as lidocaine suppressed the triggered arrhythmias at clinically relevant doses. Similar doses have been demonstrated to suppress intraventricular conduction in the same experimental model. These results suggest that the Na+ channel inhibition by class I drugs is an effective pharmacological intervention for suppressing Ca2+ overload, which may provide a rationale for the short-term use of class I drugs against the triggered arrhythmias in clinical practice.

Relationship between phosphodiesterase inhibition induced by several Kampo medicines and smooth muscle relaxation of gastrointestinal tract tissues of rats.

Given a lack of information, we assessed the effects of Kampo medicines: Dai-saiko-to, Tsu-do-san, San'o-shashin-to, and Sairei-to, which have been used for various gastrointestinal diseases, on the phosphodiesterase activity and smooth muscle tone of the gastrointestinal tract. Clinically relevant concentrations of each Kampo extract (0.1 - 1 mg/ml) decreased the phosphodiesterase activity as well as smooth muscle tone. The extent of phosphodiesterase inhibition as well as smooth muscle relaxation by these Kampo extracts was prominent for the lower gastrointestinal tract. Also, there was a good correlation between the extents of drug-induced phosphodiesterase inhibition and smooth muscle relaxation, indicating the presence of their causal link. These results may partially provide the basis for understanding the mechanism of the clinical utility of Kampo extracts in gastrointestinal tract diseases.

Comparison of the in vivo electrophysiological and proarrhythmic effects of amiodarone with those of a selective class III drug, sematilide, using a canine chronic atrioventricular block model.

Amiodarone effectively blocks both the sodium and calcium channels and beta-adrenoceptors, in addition to blocking several potassium currents including IKr, IKs, Ito, IK1, IKACh and IKNa. The incidence of clinical torsade de pointes (TdP) associated with amiodarone has been reported to be low and the present study compared the proarrhythmic potential of amiodarone with that of a selective IKr channel blocker, sematilide, using a canine chronic atrioventrucular block model. Amiodarone or sematilide (3 and 30 mg/kg; n=4 for each group) was administered orally without anesthesia under continuous ECG monitoring. Both drugs prolonged the QT interval, although the onset was faster for sematilide. The high dose of sematilide induced TdP in 3 of 4 animals, which caused their death, but neither the low dose of sematilide nor the 2 dosages of amiodarone induced lethal ventricular arrhythmias. These results suggest that IKr channel inhibition by amiodarone with its additional ion channel blocking action may contribute to the prevention of TdP.

Cardiac effects of salvia miltiorrhiza/dalbergia odorifera mixture, an intravenously applicable Chinese medicine widely used for patients with ischemic heart disease in China.

Chronotropic, inotropic and coronary vasodilator actions of the clinically available ampoule preparation of Salvia miltiorrhiza/Dalbergia odorifera mixture were examined using canine isolated, blood-perfused heart preparations. The mixture slightly decreased the sinoatrial rate and significantly increased coronary blood flow, but hardly affected the developed tension of the papillary muscle. The effect on coronary blood flow was induced by at least a 10-fold smaller dose than that which induced the chronotropic effect. These results were quite similar to those of a typical calcium channel blocker, verapamil, used in a previous study, suggesting that the Salvia miltiorrhiza/Dalbergia odorifera mixture may have potential as an anti-anginal drug.

Cardiac effects of clinically available Kampo medicine assessed with canine isolated, blood-perfused heart preparations.

Cardiac effects of 10 kinds of clinically available Kampo medicines were investigated: Kakkon-to (TJ-1), Dai-saiko-to (TJ-8), Boi-ogi-to (TJ-20), Chorei-to (TJ-40), Rokumi-gan (TJ-87), Tsu-do-san (TJ-105), Gosha-jinki-gan (TJ-107), San'o-shashin-to (TJ-113), Sairei-to (TJ-114) and Inchin-gorei-san (TJ-117). Chronotropic and inotropic effects were studied using canine isolated, blood-perfused heart preparations, while subcellular mechanisms were analyzed by measuring the drug-induced changes of the adenylate cyclase activity in the canine ventricular membrane preparation. Intracoronary injections of TJ-1, TJ-20, TJ-105 and TJ-113 increased the sinoatrial rate and developed tension of papillary muscle in a dose-related manner, which was significantly attenuated by the pretreatment of the preparations with beta-blocker propranolol. Meanwhile, the other extracts hardly affected these parameters. TJ-1, TJ-20 and TJ-113 increased the adenylate cyclase activity in a dose-related manner, but their potency was significantly less compared with that by an equivalent concentration of isoproterenol. Moreover, TJ-105 did not increase the adenylate cyclase activity. These results suggest that the positive chronotropic and inotropic effects of TJ-1, TJ-20, TJ- 105 and TJ-113 may be exerted through the direct stimulation of the beta-adrenoceptor and/or the norepinephrine release from the postganglionic nerve terminals in the heart.