Moving towards better predictors of drug-induced Torsade de Pointes. 2-3 November 2005, Crystal City, Virginia, USA

Exposure-response analysis of drug-induced QT interval prolongation in telemetered monkeys for translational prediction to human

INTRODUCTION

The preclinical in vivo assay for QT prolongation is critical for predicting torsadogenic risk, but still difficult to extrapolate to humans. This study ran preclinical tests in cynomolgus monkeys on seven QT reference drugs containing the drugs used in the IQ-CSRC clinical trial and applied exposure-response (ER) analysis to the data to investigate the potential for translational information on the QT effect.

METHODS

In each of six participating facilities in the J-ICET project, telemetered monkeys were monitored for 24 h following administration of vehicle or 3 doses of test drugs, and pharmacokinetic profiles at the same doses were evaluated separately. An individual rate-corrected QT interval (QTca) was derived and the vehicle-adjusted change in QTca from baseline (∆∆QTca) was calculated. Then the relationship of concentration to QT effect was evaluated by ER analysis.

RESULTS

For QT-positive drugs in the IQ-CSRC study (dofetilide, dolasetron, moxifloxacin, ondansetron, and quinine) and levofloxacin, the slope of the total concentration-QTca effect was significantly positive, and the QT-prolonging effect, taken as the upper bound of the confidence interval for predicted ∆∆QTca, was confirmed to exceed 10 ms. The ER slope of the negative drug levocetirizine was not significantly positive and the QTca effect was below 10 ms at observed peak exposure.

DISCUSSION

Preclinical QT assessment in cynomolgus monkeys combined with ER analysis could identify the small QT effect induced by several QT drugs consistently with the outcomes in humans. Thus, the ER method should be regarded as useful for translational prediction of QT effects in humans.

Pharmacokinetic-pharmacodynamic modelling of the effect of Moxifloxacin on QTc prolongation in telemetered cynomolgus monkeys

INTRODUCTION

Delayed ventricular repolarisation is manifested electrocardiographically in a prolongation of the QT interval. Such prolongation can lead to potentially fatal Torsades de Pointes. Moxifloxacin is a fluoroquinolone antibiotic which has been associated with QT prolongation and, as a result, is recommended by the regulatory authorities as a positive control in thorough QT studies performed to evaluate the potential of new chemical entities to induce QT prolongation in humans. The sensitivity of the cynomolgus monkey as a quantitative preclinical predictor of the PK-QTc relationship is discussed.

METHODS

Cardiovascular monitoring was performed in the telemetered cynomolgus monkey for 22 h following oral administration of Moxifloxacin (10, 30 and 90 mg/kg) or placebo. QTc was derived using an individual animal correction factor (ICAF): RR-I = QT-I--(RR-550)* (IACF). A PKPD analysis was performed to quantify the increase in placebo-adjusted QTc) elicited by administration of Moxifloxacin. In addition, the rate of onset of hERG channel blockade of Moxifloxacin was compared to Dofetilide by whole cell patch clamp technique in HEK-293 cells stably expressing the hERG channels.

RESULTS

Moxifloxacin induced a dose dependent increase in QTc). A maximum increase of 28 ms was observed following administration of 90 mg/kg Moxifloxacin. The corresponding maximum free systemic exposure was 18μM. Interrogation of the PK-QTc relationship indicated a direct relationship between the systemic exposure of Moxifloxacin and increased QTc. A linear PKPD model was found to describe this relationship whereby a 1.5 ms increase in QTc was observed for every 1 μM increase in free systemic exposure.

DISCUSSION

The exposure dependent increases in QTc observed following oral administration of Moxifloxacin to the cynomolgus monkey are in close agreement with those previously reported in human subjects. A direct effect linear relationship was found to be conserved in both species. As a result of the quantitative agreement in both species, the utility of the telemetered cynomolgus monkey as a preclinical predictor of QTc) prolongation is exemplified. Furthermore, the rate of onset of hERG channel blockade observed in patch clamp offers a mechanistic insight into the relative rates of channel blockade observed in vivo with both Moxifloxacin and Dofetilide.

The hERG K+ channel: target and antitarget strategies in drug development

The human ether-à-go-go related gene (hERG) K+ channel is of great interest for both basic researchers and clinicians because its blockade by drugs can lead to QT prolongation, which is a risk factor for torsades de pointes, a potentially life-threatening arrhythmia. A growing list of agents with "QT liability" have been withdrawn from the market or restricted in their use, whereas others did not even receive regulatory approval for this reason. Thus, hERG K+ channels have become a primary antitarget (i.e. an unwanted target) in drug development because their blockade causes potentially serious side effects. On the other hand, the recent identification and functional characterization of hERG K+ channels not only in the heart, but also in several other tissues (e.g. neurons, smooth muscle and cancer cells) may have far reaching implications for drug development for a possible exploitation of hERG as a target, especially in oncology and cardiology.

Preclinical cardiac safety assessment of drugs

The heart is a frequent site of toxicity of pharmaceutical compounds in humans, and when developing a new drug it is critical to conduct a thorough preclinical evaluation of its possible adverse effects on cardiac structure and function. Changes in cardiac morphology such as myocardial necrosis, hypertrophy or valvulopathy are assessed in regulatory toxicity studies in laboratory animals, although specific models may be needed for a more accurate detection of the risk. The potential proarrhythmic risk of new drugs is a major subject of concern and needs to be fully addressed before treatment of volunteers or patients takes place. In vitro assays are conducted to determine the effects on cardiac ion channels, in particular I(Kr) potassium channel antagonism. Prolongation of the QT interval is assessed in vivo, generally in telemetered dogs. Together, these two tests are considered to detect most arrhythmic drugs. The results of this core battery can be refined by additional studies, in particular assays on isolated cardiac tissues determining changes in cardiac action potential duration, shape and variability over time. Triggering of arrhythmia is assessed in hypokalaemic dogs with artificially created bradycardia, or in vitro in isolated whole hearts. The proarrhythmic risk of the new compound is then evaluated by integrating the results of these different tests. Drug adverse effects on cardiac electrophysiological function, in particular impulse formation and conduction, are evaluated through changes in ECG, generally recorded in dogs, pigs or monkeys. Changes in cardiac contractility occurring either as a primary effect of the drug on cardiac function or as a consequence of cardiac lesions should also be carefully assessed. In telemetered or anaesthetised animals, cardiac contractility is evaluated by measurement of left ventricular pressure and its first derivative over time. Echocardiography allows non-invasive measurement of drug-induced changes in ventricular wall movements and cardiac haemodynamics indicative of effects on contractility. In conclusion, a reliable and accurate evaluation of the cardiac safety of a new pharmaceutical agent is based on the results of in vitro tests, with overall moderate to high throughput, and in vivo experiments assessing the effects of the drug on the heart in its physiological environment. The specific sensitivities of the animals used in these assays to cardiac adverse effects should also be considered. The final evaluation of the cardiac risk is therefore based on an integrated analysis of the results from a battery of tests.

Using pharmacokinetic/pharmacodynamic modelling in safety pharmacology to better define safety margins: a regional workshop of the Safety Pharmacology Society

This meeting was convened to encourage the incorporation of empirical and mechanism-based pharmacokinetic/pharmacodynamic (PK/PD) modelling into safety pharmacology to improve the predictability of nonclinical investigations for human outcomes. These technologies make use of mathematical expressions relating measured variables to derive essential parameters for describing responses and predicting the behaviour of biological systems to a drug. Hence, empirical PK/PD modelling is intended to define the in vivo interrelationship between three basic entities: time; drug concentrations; and drug effects. The most widely applied equation relating drug bioresponses to plasma concentrations is the Hill sigmoidal E(max) model, which allows the calculation of drug potency (EC(50)) and intrinsic activity (E(max)). However, since the latter parameters depend on attributes of the drug and on the biological system itself, this approach can fail to accurately foretell drug concentration-effect behaviour, particularly between species. A particular phenomenon of PK/PD analysis is hysteresis, which refers to the delay of the bioresponse time-course with respect to exposure time-course, as this provides valuable information on the direct or indirect nature of the drug mechanism of action. The application of these concepts to the examination of the QT interval prolongation produced by dofetilide was discussed. A development surmounting the limitations of empirical PK/PD models is mechanism-based PK/PD modelling because its toolkits integrate specific mathematical expressions replicating the drug (e.g., affinity, intrinsic efficacy), and the physiological system (e.g., nonlinear, time-dependent, transduction processes), properties that play a crucial role in the cascade of biological events culminating in bioresponses. The usefulness of this approach was illustrated by a thorough analysis of nonclinical respiratory depressant and antinociceptive data on buprenorphine and fentanyl for successfully predicting the human safety and efficacy of these analgesic agents. Thus, PK/PD models can be viewed as in silico clones of drug and biological system activities that provide high-level knowledge that can avoid inappropriate attrition, and hasten the progress, of novel drugs, along the entire critical path of pharmaceutical development.

Simple-to-use, reference criteria for revealing drug-induced QT interval prolongation in conscious dogs

Electrocardiogram (ECG) QT interval prolongation produced by drugs in certain animal models is currently believed to be predictive of cardiac proarrhythmic effects in humans. For this reason, nonclinical assessment of the effects of novel drugs on cardiac repolarization is a regulatory prerequisite for progressing such agents to clinical evaluation. The present investigation was carried out to develop reliable, simple-to-use reference criteria for identifying individual animals as responders to drugs that prolong the QT interval. ECG were recorded for 30 s at 0 (8 am), 2, 4, 6 and 24 h in 6 trained, conscious, beagle dogs during 5 control experimental sessions. QT intervals were measured and corrected for heart rate by applying the Van de Water algorithm (QTc). The maximal (QTc(max)) and minimal (QTc(min)) values of QTc observed in each of the five control recording sessions were noted. Two reference (R) criteria were used to designate an individual animal as a responder to drug treatment: 1) QTc(maxR) which was obtained by adding 10 ms to the largest value of QTc(max) observed during the five control recording sessions and 2) (QTc(max)-QTc(min))(maxR) which was obtained by increasing by 50% the largest of the (QTc(max)-QTc(min)) values [(QTc(max)-QTc(min))(max)] observed in the 5 control recording sessions. The sensitivity and reliability of these criteria were tested by determining QTc intervals before and 2, 4, 6 and 24 h after placebo or quinidine (200, 400 and 800 mg p.o. per animal). The reference values of QTc(maxR) and (QTc(max)-QTc(min))(maxR) for the various dogs ranged from 246 to 270 ms and from 15 to 19.5 ms, respectively. The number of dogs responding to treatment (T: quinidine at 200, 400 and 800 mg, p.o. per animal) with a QTc(maxT) and/or a (QTc(max)-QTc(min))(maxT) equal to or greater than the respective reference values was, respectively, 1/6, 3/6 and 5/6 dogs. Additionally, the number of responders correlated well with the concentration of free quinidine in the plasma. In conclusion, this investigation succeeded in establishing reliable, reference criteria for individual dogs despite the intrinsic daily variation of QTc interval. The application of these criteria allowed identifying individual animals responding to quinidine with delayed cardiac repolarization.

Application of a probabilistic method for the determination of drug-induced QT prolongation in telemetered cynomolgus monkeys: effects of moxifloxacin

INTRODUCTION

Moxifloxacin is the most widely used positive reference agent in clinical cardiac repolarization safety studies, but it has not been characterized in the cynomolgus monkey. This important experimental animal species exhibits pronounced heart rate variability, complicating the temporal evaluation of QT interval data.

METHODS

Digitized epicardial ECGs and aortic blood pressures were collected for 20 h in telemetered cynomolgus monkeys (n=6) following the administration of either vehicle or moxifloxacin (10 or 50 mg/kg, p.o.). Moxifloxacin plasma concentrations were determined 4 h postdose. ECG intervals were analyzed by computerized algorithms. Individual probabilistic QT rate-corrections (QTc) were derived from the slopes of predose log-transformed QT-RR data where each QT value was the mean of >250 beats/RR increment. The resulting QTc was used to determine the repolarization effects of moxifloxacin, expressed as the placebo-adjusted change in QTc (DeltaQTc), and as the integrated response from 0 to 12 h (AUC(0-->12)) postdose.

RESULTS

No DeltaQTc effect was produced by 10 mg/kg moxifloxacin. However, moxifloxacin (50 mg/kg; 5.86+/-0.5 microg/mL C(max)) significantly prolonged the RR interval by 50 to 112 ms from 3.5 to 7.5 h postdose and DeltaQTc by >or=7.2 ms from 1.83 to 9.17 h, with a maximal DeltaQTc effect of +26.4 ms. No notable effects on either systemic blood pressure or body temperature occurred with either dose.

DISCUSSION

Probabilistic QT rate-corrections appear to have eliminated the confounding effects of heart rate, provided for a stable QTc baseline, and enabled the demonstration of an exposure-dependent QTc prolongation by moxifloxacin. The duration and magnitude of the QTc effect paralleled moxifloxacin pharmacokinetics, and C(max) values were similar to those achieved clinically in thorough QT/QTc studies. Thus, novel probabilistic QT rate-corrections may offer highly robust assessments of repolarization risk in both nonclinical and clinical investigations.

Novel probabilistic method for precisely correcting the QT interval for heart rate in telemetered dogs and cynomolgus monkeys

INTRODUCTION

QT intervals are not regulated on a beat-to-beat cadence, but are strongly influenced by the preceding heart rate history (hysteresis). ECG sampling, when performed over sufficiently long periods, results in the detection of ranges of different QT values for each discrete RR interval. Given the potential impact of QT hysteresis in QT interval rate-correction procedures, we hypothesized that, physiologically, the QT interval exists as a probabilistic variable where the exact value corresponding to any RR interval is precisely estimated from the associated QT population.

METHODS

Digital ECGs were collected for 18-21 h in telemetered dogs (n=7) and cynomolgus monkeys (n=7) employing epicardial ECG leads for accurate T(end) detection, and analyzed by computerized algorithms. Descriptive statistics were calculated for raw QT values in 10 ms RR increments. Individual rate-corrected QT (QTc) formulae were derived from the slopes of log-transformed QT-RR data where each QT point was the mean of >250 beats/RR increment. The aptness of this QTc model was assessed by residual analysis.

RESULTS

Beat-to-beat ECG analysis demonstrated that for all discrete cycle lengths, the associated raw QT intervals were normally distributed populations, spanning approximately 30-40 and 45-100 ms in the dog and cynomolgus monkey, respectively. In both species, QTc was stable (< or =5 ms variation) over all physiological RR intervals.

DISCUSSION

The probabilistic treatment of raw QT interval populations natively associated to any RR interval provides hysteresis-free raw QT estimates which can be accurately modeled, allowing the derivation of a precise QTc value. Previous unawareness of the probabilistic nature of the QT interval explains the historical failure of numerous QT rate-correction formulae to correctly solve this scientific issue. Importantly, QT distribution analysis has the potential to provide, for the first time, a universal and sensitive method for QT heart rate-correction, providing a robust method for nonclinical and clinical cardiac safety investigations of repolarization delay.

Blinded validation of the isolated arterially perfused rabbit ventricular wedge in preclinical assessment of drug-induced proarrhythmias

BACKGROUND

The development of preclinical models with high predictive value for the identification of drugs with a proclivity to induce Torsade de Pointes (TdP) in the clinic has long been a pressing goal of academia, industry and regulatory agencies alike. The present study provides a blinded appraisal of drugs, in an isolated arterially-perfused rabbit ventricular wedge preparation, with and without the potential to produce TdP.

METHODS AND RESULTS

Thirteen compounds were tested for their potential for TdP using the rabbit left ventricular wedges. All investigators were blinded to the names, concentrations and molecular weights of the drugs. The compounds were prepared by the study sponsor and sent to the investigator as 4 sets of 13 stock solutions with the order within each set being assigned by a random number generator. Each compound was scored semi-quantitatively for its relative potential for TdP based on its effect on ventricular repolarization measured as QT interval, dispersion of repolarization measured as T(p-e)/QT ratio and early afterdepolarizations. Disclosure of the names and concentrations after completion of the study revealed that all compounds known to be free of TdP risk received a score of less or equal to 0.25, whereas those with known TdP risk received a score ranging from 1.00 to 7.25 at concentrations less than 100X their free therapeutic plasma C(max).

CONCLUSIONS

Our study provides a blinded evaluation of the isolated arterially-perfused rabbit wedge preparation demonstrating both a high sensitivity and specificity in the assessment of 13 agents with varying propensity for causing TdP.