Romano-Ward syndrome

Excessive prolongation of the QT/QTc interval creates an electrophysiological environment that predisposes the myocardium to torsade de pointes, a polymorphic ventricular tachyarrhythmia that can progress to ventricular fibnllation and sudden cardiac death. Identification of QT/QTc prolongation liability is therefore an important objective of contemporary drug development programs. Nonclinical safety pharmacology studies are useful in early identification of this cardiac safety problem and in preliminary risk assessment. Specialized clinical pharmacology studies play a critical role in characterizing the magnitude, time course, dose dependency, and concentration relationship of drug-induced QT/QTc prolongation, as well as guiding the intensity of electrocardiogram (ECG) safety evaluations in subsequent trials. The reading, analysis, and interpretation of the ECG data acquired through these studies are complex issues that present many challenges to the pharmaceutical industry and regulatory authorities alike.

Learning Objectives

Upon completion of this article, participants should be able to

* Design and conduct clinical trials in which ECG safety, particularly QT/QTc prolongation liability, is assessed

* Analyze and interpret ECG data, particularly QT/QTc interval data, from clinical tnals

* Implement clinical research strategies for ECG assessment and analysis that will meet regulatory expectations for new drug submissions

Target Audience

This article is designed for clinical research scientists, physicians, and other staff; pharmaceutical physicians/medical directors, biostatisticians, data managers, regulatory affairs professionals, drug safety/ surveillance personnel, decision makers working in cardiac drug safety, drug regulators, clinical investigators, and consultants in cardiology and clinical pharmacology.

Key Words

Electrocardiograms (ECGs); QT/QTc prolongation; Torsade de pointes; Ventricular repolarization; Cardiac safety; Clinical trials; Design; Analysis; Interpretation; Drug regulation.

INTRODUCTION

The QT interval of the surface electrocardiogram (ECG) consists of the QRS complex, which represents depolarization within the His-Purkinje system and ventricles and the ]T interval, which reflects ventricular repolarization. The QT interval is measured from the initiation of the QRS complex to the termination of the T wave. Because of its inverse relationship to heart rate, the measured QT interval is routinely transformed by various heart rate correction formulas into a variable known as the corrected QT interval (QTc), which is intended to be independent of heart rate.

Excessive prolongation of the QT/QTc interval creates an electrophysiological environment that is conducive to torsade de pointes, a polymorphic ventricular tachyarrhythmia. Although often asymptomatic, torsade de pointes can result in syncope or progress to ventricular fibrillation and sudden cardiac death. Torsade de pointes appears on the ECG as continuous twisting of the QRS complex around the isoelectric line.

Risk factors for QT/QTc prolongation and torsade de pointes include, but are not limited to, the following (1-5):

* Female gender

* Age 65 years or older

* Presence of genetic variants affecting cardiac ion channels or associated regulatory proteins, especially those responsible for congenital long QT syndrome (LQTS; eg, Romano-Ward syndrome, Jervell and Lange-Nielson syndrome, Andersen syndrome)

* Cardiac disease (eg, myocardial ischemia or infarction, congestive heart failure, left ventricular hypertrophy, cardiomyopathy)

* History of arrhythmias (especially ventricular arrhythmias, atrial fibrillation, or recent conversion from atrial fibrillation)

* Bradycardia

* Obesity

* Acute neurological events (eg, intracranial or subarachnoid hemorrhage, stroke, intracranial trauma)

* Electrolyte disturbances (eg, hypokalemia, hypomagnesemia, hypocalcemia)

* Nutritional deficits (eg, eating disorders)

* Diabetes mellitus

* Autonomia neuropathy

* Impaired drug elimination (eg, renal or hepatic dysfunction or phenotypic poor metabolizers if relevant to the elimination of the drug)

Drugs that have been withdrawn, discontinued, or denied marketing authorization in one or more regulatory jurisdictions because of QT/QTc interval prolongation or associated proarrhythmia and sudden death include astemizole, cisapride, grepafloxacin, levacetylmethadol, lidoflazine, mesoridazine, prenylamine, sertindole, sparfloxacin, terfenadine, terodiline, and thioridazine.

HISTORY OF REGULATORY GUIDANCE

The past several years have witnessed a series of regulatory initiatives aimed at providing guidance on the assessment of QT/QTc interval prolongation liability during drug development:

1. CPMP Points to Consider: The Assessment of the Potential for QT Interval Prolongation by Non-cardiovascular Medicinal Products (December 17,1997). Available at: http://www.emea.eu.int/pdfs/human/swp/ 098696en.pdf. Accessed August 29, 2005.

2. Health Canada Draft Guidance Document: Assessment of the QT Prolongation Potential of Non-antiarrhythmic Drugs (March 15, 2001).

3. Food and Drug Administration/Health Canada Preliminary Concept Paper: The Clinical Evaluation of QT/QTc Interval Prolongation and Proarrhythmic Potential for Non-antiarrhythmic Drugs (November 15, 2002). Available at: http://www.hc-sc.gc.ca/ dhp-mps/alt_formats/hpfb-gpsa/pdf/prodpharma/ qt_concep_e.pdf. Accessed August 29, 2005.

4. International Conference on Harmonisation Step 4 Guideline: The Non-clinical Evaluation of the Potential for Delayed Ventricular Repolarization (QT Interval Prolongation) by Human Pharmaceuticals (May 12, 2005); available at http://www.ich.org.

5. International Conference on Harmonisation Step 4 Guideline: The Clinical Evaluation of QT/QTc Interval Prolongation and Proarrhythmic Potential for Non-antiarrhythmic Drugs (May 12, 2005); available at http://www.ich.org.

The purpose of this article is to examine concepts arising from the guidance development initiatives, as well as considerations emerging from regulatory experience with QT/QTc data in New Drug Submissions.

MECHANISMS UNDERLYING QT/QTc PROLONGATION

The cardiac action potential is a pattern of electrical activity generated within the myocytes by a series of currents resulting from the passage of ions across the cell membrane through voltagegated ion channels and ion pumps.

The ventricular action potential in humans consists of five successively activating phases:

* Phase O: Upstroke of the action potential resulting from the rapid, transient Na+ influx current I^sub Na^ (depolarization phase)

* Phase 1: Termination of the upstroke of the action potential resulting from the transient K+ efflux current 1^sub to^ (early repolarization phase)

* Phase 2: Plateau of the action potential resulting from the influx of Ca2+ through L-type Ca^sup 2+^ channels

* Phase 3: Sustained downward stroke of the action potential mediated by the rapid and slow components of the delayed rectifier outward K+ current I^sub gr^ and 1^sub Ks^, respectively (late repolarization phase)

* Phase 4: Resting potential maintained by the inward rectifier K+ current 1^sub K1^

Many drugs result in QT/QTc interval prolongation (6). An abbreviated list of such drugs is presented in Table 1. In the majority of cases, drug-induced QT/QTc prolongation appears to result from blockade of the IKr ion channel responsible for the rapid component of the delayed rectifier outward potassium current that brings about return of the ventricular action potential to the resting membrane potential. In humans, the pore-forming α-subunit of the 1^sub Kr^ channel is encoded by KCNH2, also known as hERG (human ether-à-go-go gene), a name that is commonly assigned to the channel as well. Drugs that delay repolarization through this mechanism include terfenadine, astemizole, cisapride, sertindole, and thioridazine (7,8).

In some cases, QT/QTc prolongation results at least in part from blockade of the sodium channels that carry the depolarizing INa current. Class IA and IC antiarrhythmics, local anesthetics, tricyclic antidepressants, and 5HT3 antagonist antiemetics have all been demonstrated to block cardiac Na+ channels (9-11). Drugs with this property tend to result in widening of the QRS complex in addition to QT/QTc prolongation. Many sodium channel blockers also interfere with hERG currents (7,8). Cocaine, a drug of abuse with dual sodium and potassium channel blocking effects, has been linked with QT/QTc prolongation and torsade de pointes in abusers (12,13).

Azimilide, bepridil, and mefloquine are drugs that appear to delay ventricular repolarization through dual blockade of IKr and lKs channels (14-16). lnhalational anesthetic gases, such as sevoflurane, halothane, and isoflurane, are associated with QT/QTc interval prolongation (17), which has been attributed to inhibition of the slowly activating delayed rectifier current I^sub Ks^ rather than I^sub Kr^ (18,19).

Drugs and toxins that delay cardiac sodium channel inactivation have been reported to provoke torsade de pointes in both humans and laboratory animals. An exploratory study of DPI 201-106, an investigational drug operating through this mechanism, was discontinued prematurely after two of three patients developed severe proarrhythmia while receiving the active treatment (20).

QT/QTc interval prolongation can also result from alterations in hERG channel trafficking, a mechanism of action that has been documented for arsenic trioxide (21) and pentamidine (22). These drugs appear to act at the level of the endoplasmic reticulum by preventing complex formation between the hERG channel and the chaperone proteins, heat shock proteins hsp70 and hsp90, such that the immature channels remain in the endoplasmic reticulum rather than being exported to the plasma membrane.

Epinephrine (23) and synthetic β^sub 2^-adrenoceptor agonists, such as terbutaline and formoterol (24), have been reported to result in QT/QTc prolongation. These effects may be mediated through multiple mechanisms, including a decrease in serum potassium concentrations (25) and inhibition of I^sub Kr^ channel activity via cyclic adenosine monophosphate (cAMP)/protein kinase A-dependent pathways (26). QT/QTc prolongation has also been associated with some natural health products containing sympathomimetic substances (27,28).

Loop and thiazide diuretics have the potential to cause QT/QTc prolongation by inducing hypokalemia or hypomagnesemia (29).

In addition to small molecular weight chemicals, several peptide hormones and peptide hormone analogues, such as oxytocin (30), vasopressin (31,32), and octreotide, as well as the protein hormone insulin (33,34), have been associated with QT/QTc prolongation or proarrhythmia after parenteral use for therapeutic purposes. The mechanisms underlying these effects have not been fully elucidated. In the case of insulin, the QT/QTc prolongation appears to result from increased sympathetic tone and decreased potassium concentrations secondary to hypoglycemia (33,34).

The gonadotropin-releasing hormone agonists and antagonists (eg, abarelix, leuprolide, and goserelin) have been reported to cause QT/QTc prolongation when used for androgen ablation in the palliative treatment of prostate cancer (35). Although the mechanism of action has not been established, androgens are thought to have a role in regulating cardiac ion channel expression (36). Some studies have also reported a link between estrogen replacement therapy and QTc prolongation in postmenopausal women (37,38).

Several naturally occurring peptide toxins can cause delayed repolarization by binding to the exterior of ion channels in the myocardium. For example, certain sea anemone toxins, such as anthopleurin A, interfere with deactivation of the cardiac Na+ channels (39), while scorpion toxin has been demonstrated to block hERG channels (40,41). Combretastatin Bl, a polyhydroxybibenzyl toxin extracted from the South African plant Combretum krausii, has been observed to suppress hERG-like currents in a cancer cell line (42).

INTENSIVE ECG SAFETY EVALUATION IN PHASE I AND II CLINICAL TRIALS

A clinical pharmacology study of adequate statistical power, dedicated to the assessment of QT/QTc prolongation liability, is an expected component of almost all new drug submissions for pharmaceuticals with systemic bioavailability. Intensive monitoring of ECG safety is expected in all phase I and II clinical trials performed prior to this dedicated ECG safety evaluation.

ROLE OF NONCLINICAL SAFETY PHARMACOLOGY STUDIES

Nonclinical safety pharmacology studies are available to assess the potential of drugs to delay ventricular repolarization (43). In the case of many QT/QTc-prolonging drugs, these studies have proved useful for hazard identification and risk assessment. Unfortunately, safety pharmacology studies have not been demonstrated to consistently detect signals for some drugs that prolong the QT/QTc interval to a small, but noteworthy, extent in humans. Interpretation of equivocal results can be difficult as there are no established thresholds signifying cause for concern.

In many drug submissions, an assay of acute effects on hERG currents in a heterologous expression system or I^sub Kr^ in isolated cardiomyocytes will be the only available in vitro test for assessment of the potential for delayed ventricular repolarization. While hERG/IKr channel blockade certainly accounts for the QT/QTc prolongation observed with many drugs, several other mechanisms have been identified as well; these are not routinely studied in the standard nonclinical safety pharmacology program.

The comparison of in vitro potency estimates with in vivo plasma concentrations presents many challenges. Binding of the drug to the glass or plastic of the apparatus or nonspecific binding to the test matrix can decrease the in vitro concentration; however, experimental verification of the actual concentration in the incubation chamber is not routinely performed (44). Low aqueous solubility or degradation of the test article can also account for failure to obtain a useful potency estimate (44). If one or more metabolites contribute to QT/QTc prolongation in vivo, in addition to or instead of the parent compound, comparisons of in vitro to in vivo concentrations could be particularly misleading.

Methodological differences can result in large discrepancies in potency estimates for the same compound between laboratories, with IC^sub 50^ values varying by two or three orders of magnitude (45). This variability may be related to differences in expression system, potassium concentrations, voltage pulse protocols, temperature, or other factors (45,46).

Controversy exists concerning the choice of free or total plasma concentrations for comparison purposes. Myocardial tissue concentrations are likely to be more relevant than plasma concentrations, especially for lipophilic drugs with high tissue penetration. A study of myocardial tissue distribution of several antipsychotic drugs in guinea pigs showed myocardium-toplasma concentration ratios ranging from 2.2 to 6.4 (47). The determination of myocardial-toplasma concentration ratios is not realistic for humans and indeed not routine even for laboratory animals. Considering the many uncertainties involved, attempts to relate in vitro potency estimates to human exposure are not considered appropriate for regulatory decision making.

Many regulators continue to be concerned about the possibility of false-negative findings in the in vivo cardiac safety pharmacology studies performed in conscious and anesthetized laboratory animals. Interspecies differences in electrophysiology, hemodynamics, and drug metabolism can limit the ability to extrapolate nonclinical ECG findings to the human situation. Tolerability problems may prove to be dose limiting or have confounding effects on the ECG end points. Other limitations include small sample size (eg, N = 3-4), single-dose protocols, and inadequate use of positive controls to validate the assays. Typically, nonclinical safety pharmacology studies have low or undefined sensitivity (eg, 10% change in the QT/QTc). Attempts to define and improve the sensitivity of these test systems are particularly encouraged.

Although nonclinical electrophysiology studies cannot substitute for a dedicated clinical pharmacology study of QT/QTc prolongation liability in humans, the results from these assays are useful in evaluating early clinical trials for the appropriateness of the proposed dose escalation steps, the ECG monitoring schedule, eligibility and discontinuation criteria, and informed consent. Mechanistic information regarding effects on ion channel function or processing will, of course, be valuable at all stages of drug development in guiding the interpretation of clinical ECG findings and adverse events suggestive of proarrhythmia.

INTENSIVE ECG SAFETY MONITORING IN CLINICAL TRIALS PRIOR TO THE QT/QTc ASSESSMENT STUDY

An intensive ECG safety evaluation in phase I clinical pharmacology studies has an important role in monitoring the safety of subjects and guiding the ECG evaluation strategy in phase II trials. The inclusion of a placebo control group with time-matched ECG recordings is critical to allow drug-related changes in the QT/QTc interval to be distinguished from those caused by diurnal variations, postprandial effects, alterations in autonomie tone, and other factors. ECG recordings should be performed at baseline and at multiple postdose time points to characterize the time course of any QT/QTc prolongation in terms of onset, peak effect, and offset. The collection of replicate ECG recordings at each time point is encouraged to minimize intrasubject variability and reduce the measurement error (eg, three or more ECGs recorded over a period of 4 min or less around each nominal time point).

Crossover study designs permit within-subject placebo comparisons, thus maximizing the statistical power for a given sample size. If a parallel group study design is used, a full day of baseline replicate ECGs should be collected at time points chosen to coincide with the scheduled on-therapy recordings.

To the extent possible, experimental conditions should be standardized, with the ECG recordings performed in supine, resting subjects. Blood samples for pharmacokinetic determinations should be collected immediately after the ECG recordings are completed. Correlations between QT/QTc changes and plasma concentrations of the parent drug and any active metabolites will be an important consideration in evaluating data from these early studies (eg, slope of the delta QTc-versus-concentration relationship). If the investigational drug has dual effects on potassium concentrations and the QT/QTc interval (eg, 32-adrenergic agonists), superimposed graphs showing the magnitude and time course of these effects can be a valuable means of exploring an association between the two phenomena.

If a potential for delayed ventricular repolarization has been detected in nonclinical safety pharmacology studies, the Investigator's Brochure should summarize these findings and discuss their possible clinical implications. The Informed Consent Form should provide an explanation of the theoretical proarrhythmic risk in lay language. This text should be retained until sufficient clinical evidence has accumulated to dismiss the potential risk with reasonable confidence (ie, completion of a dedicated ECG assessment study with the statistical power to detect a mean increase in the QTc interval of 5 ms or less). If treatment-emergent QT/QTc prolongation or adverse events suggestive of proarrhythmia become apparent at any stage of clinical drug development, the Investigator's Brochure and Informed Consent Form should be amended to reflect these findings.

DEDICATED CLINICAL PHARMACOLOGY STUDIES OF QT/QTc PROLONGATION LIABILITY

A clinical pharmacology study of adequate sensitivity, dedicated to evaluating effects on cardiac repolarization, should be performed at a time point in the drug development program when the therapeutic dose range has been established and the pharmacokinetics characterized in terms of half-life, metabolic profile, and the mean and range C^sub max^ values for analytes of interest under conditions of impaired elimination, such as maximal metabolic inhibition.

While the membership of an investigational drug in certain structural (eg, macrolide, fluoroquinolone) or pharmacological (eg, antipsychotics) classes may elicit heightened suspicions regarding its QT/QTc prolongation potential, historical experience with a drug class should not be used as an argument for dismissing the possibility of an effect. Noteworthy examples of substances that exhibit marked differences in QT/QTc prolongation liability, despite having a close structural relationship, include terfenadine and fexofenadine (48). While efforts are currently ongoing to develop in silico methods to assess QT/QTc prolongation potential on the basis of chemical structure, additional experience is needed to determine the predictive value of these tests (49).

Brevity in terms of duration of use or duration of action of the drug does not provide a defensible basis for exemption from a meticulous ECG safety evaluation. The risk of proarrhythmia seems to be particularly high for parenteral dose forms of many commonly used drugs, such as erythromycin (50,51), droperidol, and haloperidol (52,53) and can occur after single doses. The peptide hormones oxytocin (30) and vasopressin (31,32) have been reported to result in QT/QTc prolongation and proarrhythmia despite elimination half-life values of only a few minutes.

SUBJECTS

Clinical pharmacology studies dedicated to the assessment of QT/QTc prolongation liability are generally performed in healthy volunteers. The use of healthy volunteers has the advantage of minimizing confounding factors that may contribute to variance and interfere with the ability to distinguish a drug-related change in the QT/ QTc interval. Furthermore, healthy volunteers are sometimes able to tolerate higher doses of the test drug than could a patient population.

Some important exceptions should be noted, however. Many drugs intended for the treatment of cancer, schizophrenia, and Parkinson's disease are too toxic or poorly tolerated to be studied in healthy volunteers. In such cases, careful monitoring in a patient population may be the only option for assessing the ECG safety.

DESIGN

Clinical pharmacology studies intended to assess QT/QTc prolongation liability should be of double-blind, randomized design. Provisions for bias control are important because the QT/QTc interval can be affected by posture, activity level, food ingestion, emotional state, and other factors that could conceivably be influenced by the research staff or the subjects if the identity of the treatments was known.

Both crossover and parallel group designs have proved useful for QT/QTc assessment studies. By having subjects serve as their own controls, crossover designs allow the desired statistical power to be achieved with a smaller sample size. In crossover studies, the end points of interest can be adjusted for placebo using timematched, within-subject comparisons, whereas for parallel group studies, time-matched placebo adjustments present challenges when the timing of the end point of interest can vary between subjects (e.g., mean change at subjectspecific C^sub max^/T^sub max^). Potential disadvantages of crossover studies include lengthy study duration, carryover effects, and limited ability to use data for subjects who discontinued prematurely without receiving placebo.

Parallel group studies can be of shorter duration than crossover studies and may be preferred in the following situations:

* Trials with a large number of treatment groups

* Drugs that require incremental titration to the target dose for reasons of tolerability

* Drugs/active metabolites with long elimination half-lives, for which lengthy time periods would be required to reach steady state or achieve washout

* Drugs that would be expected to have carryover effects (eg, enzyme induction)

* Studies performed in patients for whom the duration of exposure to a nontherapeutic clinical trial situation should be minimized for ethical considerations

Use of a parallel group design does, however, necessitate the enrollment of a much larger sample size to achieve the same statistical power as a crossover study (54). In some cases, a reasonable alternative to consider might be a balanced incomplete crossover design, in which subjects are randomized to receive placebo and a subset of the active treatments, under crossover conditions.

Both placebo and positive control treatment arms should be included in a QT/QTc assessment study. The positive control should be an agent of well-characterized QT/QTc prolongation liability that can be used to establish the validity of the study and define its sensitivity in terms of the end point of interest. The sensitivity target is currently defined by the International Conference on Harmonisation (ICH) guideline as a mean increase in the QTc interval of approximately 5 ms. Positive controls that routinely produce relatively large signals (eg, mean increase of 10 ms or longer) may have some value in establishing the validity of new methodologies or in providing a context for the effects of the investigational drug but do not provide a convincing demonstration of assay sensitivity. Furthermore, it should be noted that a lower-than-expected effect size for a particular positive control treatment (eg, an observed maximum mean increase of 5-6 ms when historical values usually range from 10 to 20 ms) would raise questions about the sensitivity of the study in question.

Desirable characteristics of a positive control treatment include the following:

* Good tolerability

* No effects on hemodynamics or level of consciousness

* No induction or inhibition of drug-metabolizing enzymes

* Low interindividual variability in the C^sub max^, T^sub max^, and area under the concentration-time curve (AUC)

* Low interindividual variability in the pharmacodynamic Tmax for QT/QTc prolongation

* Relatively rapid elimination of the parent drug and metabolites, such that washout periods in a crossover study need not be inordinately long

* A fairly sustained effect on QT/QTc prolongation, such that the peak effect will not be easily missed

* A dose form that is amenable to blinding

Research initiatives aimed at characterizing useful oral, intramuscular, intravenous, and inhalational positive control treatments are encouraged. As cumulative effects on QT/QTc prolongation have been reported over the course of repeat-dose administration for many drugs, different positive control treatments may be needed for single- and multiple-dose studies. If a single-dose positive control treatment is to be used in a multiple-dose study of an investigational drug, a placebo should be administered on the intervening study days so that the duration of time between the baseline and on-treatment ECG recordings is the same for the positive control as for the other treatment groups or periods.

If an investigational drug belongs to a structural or pharmacological class that has been associated with QT/QTc prolongation, a reference agent selected from other members of the same class is recommended to permit a comparison of effect sizes, preferably at equipotent therapeutic/supratherapeutic doses (55-57).

If QT/QTc prolongation liability has already been identified through intensive ECG monitoring in standard exploratory trials, the design of the dedicated ECG assessment study could be modified, with the sample size calculation based on the effect sizes observed for the investigational drug and its reference compound(s) in previous trials. A positive control for the purpose of defining study sensitivity might be considered dispensable under these circumstances.

For a given subject, repeat ECG measurements should be performed using the same machine. The use of a semipermanent skin marker is valuable to ensure consistent placement of the leads. The QT/QTc interval should be computed as the average of three or more consecutive complexes from replicate (three or more) ECG recordings at each nominal time point to increase the precision of measurements.

The conditions of ECG collection should be standardized to the extent possible (eg, posture, activity level, timing in relation to meals). As sleep results in QT/QTc prolongation, it is important to ensure that subjects are conscious during the ECG recording. The collection of blood samples in close temporal proximity to the ECGs is encouraged to permit an exploration of possible pharmacokinetic-pharmacodynamic relationships; however, such samples should be drawn after the ECG recordings have been completed to avoid the possible confounding effects of pain or anxiety related to venipuncture.

DOSE SELECTION

Clinical pharmacology studies of QT/QTc prolongation liability should assess the maximum recommended therapeutic dose of the drug. Supratherapeutic doses should also be studied, provided, of course, that safety and tolerability are not prohibitive. If possible, the supratherapeutic doses should produce plasma concentrations that exceed the upper limit of the range of Cmax values observed in subpopulations who may have reduced capacity to eliminate the drug (eg, elderly, poor metabolizers, those with renal or hepatic impairment) or in cases of drug-drug interactions (eg, maximal inhibition of a major metabolic pathway). As patients and physicians will often titrate dosage beyond the maximum recommended dose in anticipation of increased efficacy, consideration should also be given to the dose size that might be used in intentional overdosing with therapeutic intent, especially for drugs used in outpatient settings. Even for parenterally administered drugs used exclusively in inpatient settings, the study of supratherapeutic concentrations may be warranted because drugs intended for use as slow intravenous infusions will occasionally be administered as rapid bolus injections in violation of prescribing information (30). In many cases, the choice of supratherapeutic dose will be limited by ethical considerations. For some orally administered drugs, saturation of gastrointestinal absorption may limit the ability to achieve supratherapeutic concentrations; for parenteral drugs, solubility considerations may impose constraints on the maximum dose that can be studied.

If the investigational drug is a sensitive substrate for a particular drug-metabolizing enzyme (eg, CYP3A4 or CYP2D6), supratherapeutic concentrations can be achieved by coadministration of a metabolic inhibitor. This approach will be suitable only if major metabolites of the parent compound do not have an important role in contributing to the QT/QTc prolongation effect. Use of pharmacological inhibitors has the possible disadvantage of introducing confounding effects on the parameters of interest, whether through effects on ion channel function or hemodynamics. Nevertheless, the use of inhibitors may be valuable for achieving supratherapeutic plasma concentrations of drugs that have limited absorption from the gastrointestinal tract or poor gastrointestinal tolerability at high doses.

Concentration-response relationships can often be adequately assessed in a study with only two dose groups; however, a minimum of three doses should be included if the dose-response relationship is to be assessed. Observation of a relatively flat dose-effect relationship in a trial that includes only two disparate doses may leave lingering questions concerning the possibility of a bell-shaped dose-response curve.

To the extent possible, doses of the investigational drug and reference agents should be matched in terms of potency (eg, maximum recommended therapeutic dose or equivalent multiples thereof). The duration of the washout period should be selected on the basis of the pharmacokinetic and pharmacodynamic properties of the investigational drug, the reference agents, and the positive control.

TIME COURSE

If the investigational drug is intended for repeat-dose use, a multiple-dose QT/QTc assessment study is encouraged, with intensive ECG monitoring under conditions of steady-state plasma concentrations of the drug and its active metabolites.

For drugs with short elimination half-lives and no active metabolites, a single-dose study might be acceptable if an appropriate scientific justification is provided and supported by pharmacokinetic data.

The time course of QT/QTc prolongation should be fully characterized over the course of a dosing interval in terms of onset, peak, and offset of effects, with the choice of time points guided by the plasma concentration-time profiles of the drug and its major metabolites. Multiple replicate ECGs should be collected at baseline, at time points throughout the dosing interval, and prior to release of the subject from the laboratory. The time points should be selected to accurately capture the maximum effects of the investigational drug, positive control, and any reference agent.

While care should be taken to schedule ECG recordings at time points that correspond to the expected time of peak plasma concentrations of the parent compound, this is not sufficient. Maximum QT/QTc prolongation is often delayed in relation to the pharmacokinetic T^sub max^ because of the contribution of active metabolites (58), delayed distribution of the parent drug or metabolites to the myocardial tissue, or effects on ion channel trafficking or expression (21,22). If QT/QTc prolongation develops progressively over the course of several days, as with arsenic trioxide (59), the time course of the effect might have to be studied beyond the attainment of steady-state plasma concentrations.

SPECIAL CHALLENGES

Some drugs present special challenges to the characterization of their ECG safety. For instance, many investigational drugs for the treatment of cancer are too toxic to study in healthy volunteers. Other therapeutic classes (eg, dopaminergic drugs for the treatment of Parkinson's disease or dopamine receptor antagonists for the treatment of schizophrenia) are so poorly tolerated by healthy volunteers that the maximal tolerated doses in phase I studies may be less than the therapeutic doses. Such drugs are often best studied for QT/QTc prolongation liability in clinical pharmacology studies conducted in the target patient population (56). Intensive ECG monitoring can also be incorporated in first-in-human studies performed in patients with cancer that are designed with both safety and efficacy objectives (60,61).

If use of a placebo control group is deemed unethical, efforts should be undertaken to collect a full day of baseline ECG recordings at time points selected to coincide with the ontherapy recordings. The on-therapy time points should be carefully chosen to capture the rising, peak, and declining phases of the plasma concentration time course of the drug over a dosing interval. The use of triplicate ECG recordings at each scheduled time point is strongly encouraged to reduce variability to the extent possible.

If the patients have been receiving prior medications that are known to cause QT/QTc prolongation (e.g., antipsychotics), it will be important to schedule a sufficiently long washout period to ensure that carryover effects do not lead to an inflated baseline QT/QTc interval that might result in underestimation of any prolongation caused by the investigational drug. Collection of baseline ECGs on multiple days may be warranted to ensure that the QT/QTc interval is relatively stable.

QT/QTc prolongation has been reported with certain synthetic opioid drugs, such as levacetylmethadol and methadone (62). Exploration of the dose-response relationship of opioid analgesics in healthy volunteer populations may be difficult, however, because of dose-limiting adverse events in subjects who are not tolerant to the side effects of these drugs. QT/QTc assessment studies in opioid-tolerant patients would present prohibitive ethical issues because discontinuation and washout of their existing treatment might result in intractable pain and severe withdrawal reactions. Intensive ECG monitoring in subjects with histories of opioid abuse would provide an opportunity to assess the QT/QTc prolongation liability of the investigational drug at high doses. Clinical pharmacology studies in former addicts are commonly performed to assess the abuse liability of new opioids.

Gonadotropin-releasing hormone antagonists (eg, abarelix) and agonists (eg, leuprolide, goserelin) have been reported to result in sustained QT/QTc prolongation when used for androgen ablation in the palliative treatment of prostate cancer (35). The collection of ECGs at several points over the days, weeks, and months following injection of these products would be of great interest to characterize the time course of these effects.

The peptide hormones oxytocin (30) and vasopressin (31,32) have been associated with proarrhythmia following rapid bolus injection. Continuous ECG monitoring during the postinjection period is advisable to characterize the time course of possible effects with peptides and other rapidly acting, short half-life, parenterally administered drugs.

PHASE II AND III CLINICAL TRIALS CONDUCTED AFTER THE QT/QTc ASSESSMENT STUDY

A standardized schedule of ECG monitoring is encouraged for phase II and III clinical trials to facilitate subsequent integrated analyses of pooled data from multiple studies. The phase II and III ECG data have the advantage of providing information from the intended patient population for whom drug-disease and drug-drug interactions may become evident. Although the density of ECG sampling in phase II and III clinical trials is seldom sufficient to provide a reliable estimate of peak effect size, valuable information can often be obtained regarding dose dependency, as well as the magnitude of effect and incidence of outliers with the investigational drug relative to standard therapies for the indication in question.

Although on-site evaluation of the ECGs will be necessary for the rapid assessment of patient safety, the ECG database for regulatory purposes should be processed by a central laboratory in which standardized conventions for the reading and interpretation of the ECGs can be observed.

When the ECG assessment study or other previous clinical trials have provided evidence of QT/QTc prolongation, consideration should be given to implementing the following exclusion criteria to avoid exposure of patients who may be at high risk of proarrhythmia:

* Baseline prolongation of the QTc interval (eg, 450 ms or longer for the Bazett-corrected QT interval)

* History or family history of congenital LQTS or Brugada's syndrome

* Electrolyte disturbances at baseline or screening (eg, hypokalemia, hypocalcemia, hypomagnesemia) or conditions predisposing the patient to electrolyte abnormalities (eg, eating disorders)

* History of syncope

* History of myocardial ischemia or infarction, congestive heart failure, left ventricular hypertrophy, cardiomyopathy, or cardiac conduction defects

* Documented history of arrhythmias (eg, ventricular arrhythmias, atrial fibrillation)

* Implanted defibrillators or pacemakers

* Concomitant use of other drugs that delay ventricular repolarization

* Concomitant use of drugs that decrease the metabolism or increase the bioavailability of the investigational drug

The degree of caution observed in selecting the inclusion and exclusion criteria for patient eligibility will be dependent on the risk-tolerance level that is considered appropriate for the indication in question.

In phase II and III clinical trials, ECG recordings are typically performed at screening to evaluate the eligibility of the patient and at baseline to obtain information on the pretherapy status of the patient to use as a point of reference for distinguishing treatment-emergent changes. The baseline measurements should be performed close to the time of active treatment. When there is cause for concern regarding QT/QTc prolongation, multiple ECGs should be recorded during the baseline visit so that the interval measurements can be computed as means from several recordings (63). The reduced variance of the resulting interval parameters will provide a better estimate of the pretherapy status of the patients, enabling on-therapy changes to be determined more accurately. Replicate determinations at each of the scheduled on-therapy ECG measurements would likewise serve to reduce variance and improve the precision of measurement when concerns exist about QT/ QTc prolongation (63).

In phase II and III trials, ECGs are typically scheduled at regular intervals during steadystate treatment. When cause for concern has been identified in the clinical pharmacology laboratory setting, careful ECG assessment will be expected in phase II and III trials, including a characterization of the dose, concentration, and time dependency over the entire range of therapeutic doses. The magnitude of QT/QTc prolongation at supratherapeutic plasma concentrations should be explored as well if not prohibited by ethical considerations. To the extent possible, the ECG recordings should be scheduled to coincide with the expected time of peak QT/QTc prolongation as determined in the clinical pharmacology studies. The collection of blood samples for pharmacokinetic determinations in close temporal proximity to the ECGs will permit an examination of the concentration-effect relationship. When strong concerns exist about QT/QTc prolongation liability, it may be advisable to have an ECG performed after the first dose or after incremental dose adjustments in addition to steady state. The collection of ECGs at the time of adverse events suggestive of proarrhythmia is particularly encouraged. ECGs should be scheduled at the end of the treatment period and after discontinuation to determine the reversibility of the effects after cessation of drug ingestion.

In some cases, a dedicated clinical pharmacology study of QT/QTc prolongation liability in the relevant patient population or a vulnerable subpopulation might be needed to accurately characterize the effect.

If the QT/QTc assessment study does not indicate cause for concern and there are no other electrophysiological, hemodynamic, or biochemical signals that raise suspicions about cardiac toxicity, then routine ECG safety monitoring, with baseline and periodic on-therapy readings, will usually suffice for subsequent trials. Under these circumstances, automated ECG interval readings, without manual overread, might be considered sufficient.

Whether administration of high supratherapeutic doses of a drug to healthy volunteers will provide an approximation of the QT/QTc prolongation that would be observed in patients with risk factors is, of course, debatable. Certainly, if testing of high doses in healthy volunteers is not possible for reasons of safety or tolerability or if supratherapeutic exposure cannot be achieved because of limited gastrointestinal absorption, the QT/QTc assessment study would be considered inconclusive and should not be used as a basis for pursuing a diminished ECG safety assessment in subsequent trials or dismissing concerns arising from the nonclinical safety pharmacology studies.

Numerous other considerations would justify more intensive ECG monitoring in therapeutic clinical trials independent of the results of the QT/QTc assessment study. Evidence of drug-induced effects on heart rate or cardiac conduction would raise concerns about potential proarrhythmic activity. Drugs that predispose patients to certain metabolic disturbances would raise similar suspicions. For example, diuretic or nephrotoxic effects can lead to electrolyte imbalances with important electrophysiological consequences for the myocardium. Some studies have reported an association between QT/QTc prolongation and the therapeutic use of certain hormones and hormone analogues, such as those mimicking or affecting the release of gonadal steroids. As the mechanism and time course of these effects have not been elucidated, intensive ECG monitoring for such drugs in therapeutic clinical trials would appear to be advisable regardless of the outcome of the QT/QTc assessment study. Concerns about potential ischemic damage caused by prothrombotic, dyslipidemic, or vasoconstrictor effects should likewise have an impact on ECG monitoring.

A high prevalence of risk factors for arrhythmia or ischemic injury in the target patient population would also affect the extent of the ECG safety evaluation in therapeutic clinical trials. Any evidence of treatment-emergent ECG abnormalities, heart rhythm disturbances, or ischemic injury from phase II or III clinical trials with the investigational agent would, of course, dictate intensified ECG monitoring in subsequent trials even if the QT/QTc assessment study did not provide evidence of cause for concern.

READING, ANALYSIS, AND INTERPRETATION OF ECG DATA FOR QT/QTc PROLONGATION LIABILITY

ECG data should be analyzed for both the uncorrected and corrected QT intervals, as well as the PR and RR intervals and the QRS duration. Various alternative ECG repolarization parameters have been suggested, such as the T-end interval (the interval between the peak and the end of the T wave), the JT interval (the difference between the QT interval and the QRS duration), and the area under the T wave, but clinical experience is presently inadequate to establish the prognostic value of these measurements.

READING ELECTROCARDIOGRAMS

Modern electrocardiographs collect the ECG waveform as a digital signal that contains encoded time and voltage information for each of 12 simultaneously recorded leads. Advantages of digital ECGs over traditional paper ECGs include the following (64,65):

* Avoidance of problems related to changes in the physical characteristics of paper caused by temperature and moisture

* Avoidance of mechanical fluctuations affecting writer outputs

* Avoidance of problems related to distortion of the image during photocopying or scanning of a paper ECG

* Greater precision of readings

* Amenability to manipulations by software applications that provide alternative modes of viewing the waveforms (eg, with/without grids, with/without annotations, superimposition of ECG waveforms from all leads to form a representative beat)

* Facilitation of standardized conventions for annotation (eg, placement of fiduciary points)

* Facilitation of analyses for new metrics currently under investigation (eg, area under T wave)

ECG measurement software is used to display the digital ECG waveform on a computer screen, on which the ECG reader can use virtual calipers to mark the fiduciary points of the ECG intervals, a process known as manual reading. ECG readings can also be accomplished using computer algorithms that provide automatic placement of the fiduciary points. Many automated methods have been criticized for their poor accuracy, especially when the T wave has low amplitude, abnormal morphology, an indistinct termination, or overlaps with a U wave. Some automated methods may, however, have the advantage of greater reproducibility than manual measurements.

When the accurate determination of ECG intervals is an important study objective, trained professionals operating from a central laboratory should perform manual measurements of the intervals or review the computerized measurements, with adjustment of the fiduciary points wherever the automated placement is judged to be inappropriate. The ECG readers should be blinded to time, treatment, and patient identity when performing interval measurements. The same reader should measure all ECGs for a given subject, as well as a constant proportion of subjects in each treatment group. Quality assurance procedures should be observed, including assessment of inter- and intrareader variability.

Efforts to develop more sophisticated and reliable algorithms for automated ECG readings are encouraged. Each automated algorithm should be validated independently for diverse subject populations and both normal and abnormal ECG waveforms. Validation studies should use positive control agents to demonstrate the ability of the algorithm to detect noteworthy drug-induced changes.

Technically defective ECGs should not be used for the generation of interval data. Care should be taken to minimize changes in heart rate during the time when the ECGs are recorded. When a heart rate change occurs, the adaptation of the QT interval is not instantaneous, a phenomenon known as QT/RR hysteresis (66). For this reason, QTc determinations should not be based on ECG waveforms collected at times of heart rate instability. Use of relatively long ECG recordings (eg, 30 s rather than the standard 10s) may be a useful strategy as this will provide more options for reading purposes when several waveforms have to be disqualified because of the QT/RR hysteresis phenomenon (67).

At present, several different conventions are in use for the choice of ECG leads on which to base QT interval readings:

* Lead II

* Lead II when measurable, with lead V2 as the second choice

* Lead where the QT interval is measurable with the greatest precision (eg, steepest slope in the descending limb of the T wave)

* Mean/median/maximum QT interval from 12 simultaneously recorded leads

* Earliest onset to latest offset among 12 simultaneously recorded leads

The measured QT interval and the magnitude of drug-induced changes will vary depending on the leads or combinations of leads used by the reading method (68). Limiting measurements to a single lead can present problems in the event of low-amplitude T waves that are difficult or impossible to read. Precision can be improved by superimposing all ECG leads on the same baseline to define the overall QT interval (67).

Different approaches are also in use for dealing with U waves. Small, discrete U waves should be excluded from measurement of the QT interval. In certain cases, however, the U wave is large and overlaps with the T wave. Some authorities would consider this type of waveform to be a T/U complex, while others would prefer to describe it as a biphasic T wave. In the presence of overlapping U waves, T wave offset has been defined as the intersection of the tangent to the steepest downward slope of the T wave with the isoelectric line or the nadir between the T and U waves. As a large U wave is considered indicative of delayed repolarization, the most conservative approach is to include the merged U wave in the determination of the QT interval (69).

Twelve-lead Holier recording is an evolving technology that has proved useful in some ECG assessment studies. At present, concerns exist about background noise and QT/RR hysteresis in studies conducted in ambulatory subjects. Studies in supine, resting subjects with 12-lead Holter recording can yield useful results. Any Holter recording equipment used in an ECG assessment study would have to be validated independently, with the use of a positive control to demonstrate the desired level of sensitivity. The QTc data obtained from 12-lead Holter recording will not necessarily correspond quantitatively to those obtained using standard 12-lead ECGs (70). For this reason, data obtained from the two technologies are not suitable for direct comparison or evaluation using the same outlier thresholds unless, of course, validation studies have been preformed to determine the equivalence of the two methods for the equipment in question.

HEART RATE CORRECTION

The QT interval has an inverse relationship to heart rate. For this reason, various formulas are used to correct the QT interval for the influence of heart rate. Ideally, a scatter plot of the derived QTc-versus-RR values should generate a horizontal linear regression line (slope = 0), indicating independence of the QTc intervals from the RR values.

For the individual-specific regression approach to be reliable, a very large set of drugfree QT-RR measurements (eg, 400 QT-RR pairs) should be available for each study participant, with the RR values covering a broad range (eg, 600 to 1000 ms) (72). The range of heart rates during active treatment should match that observed during the drug-free period. Sponsors using this approach should be prepared to demonstrate the extent to which these criteria were met in their studies. In the setting of a clinical pharmacology laboratory, in which ECGs are collected from resting subjects, the range of heart rate values is often not adequate to support this heart rate correction method.

The RR bin method of controlling for heart rate involves distributing QT values according to their preceding RR interval into "bins" encompassing a predefined range (73). For example, the QT1000, determined from the 995- to 1004-ms RR bin, is an estimate of the QT interval at 60 beats per minute. This approach is not amenable to the examination of time course or concentration-effect relationships and is therefore suitable only as an auxiliary analysis.

For any given study, the QT data set should be corrected for heart rate using several different methods. The end points of interest should be computed for each of the resulting QTc data sets. The methods used to correct or control for heart rate can lead to considerable differences in the computed end points (70,74-76). For this reason, the choice of heart rate correction approach should affect the thresholds used for eligibility and discontinuation criteria and the interpretation of outlier analyses. Explanations should be offered for any discrepancies in results between the different heart rate corrections.

ANALYSES OF CENTRAL TENDENCY FOR QT/QTc DATA

For all treatment groups/periods, the QT/QTc value at each time point should be expressed in terms of a mean, a mean change from baseline ΔQT, and a standard deviation. If spontaneous variability has been adequately controlled, changes from baseline in the placebo group should be very small. To facilitate assessment of a possible time course relationship, both tabular and graphical presentations of these data should be provided.

The investigational drug will be compared to the placebo and active control treatments in terms of the differences of means at serial postdose time points. A small mean increase in the QT/QTc interval, which appears not be clinically significant in itself, may nonetheless signal an enhanced risk with the investigational drug if not matched by a corresponding change in the placebo control group. The effect size is determined as the difference between the timematched, baseline-adjusted QT/QTc values for the drug and placebo treatments (ie, ΔΔQT). This point estimate should be accompanied by the two-sided 90% confidence interval or the one-sided 95% confidence interval upper bound.

Many different methods are currently in use for computing the baseline QT/QTc value, including, but not limited to, the following:

* Baseline values at time points scheduled to match the on-treatment recordings for each treatment group/period

* Average of all baseline values for the treatment group/period or all periods

* Predose value for the treatment group/period

* Average of predose values for all treatment periods

* Average of all values on the placebo day

In a crossover study, period-specific baseline values are needed to provide information on possible carryover effects. In a parallel group study, a full day of time-matched replicate baseline values may be particularly valuable to account for within-subject diurnal patterns. Use of a predose baseline or an average of all baseline values assumes that there is no diurnal pattern in QT/QTc changes, while use of timematched baselines assumes that diurnal variations in the QT/QTc are reproducible from day to day within individuals.

In some cases, correction of the same data set using a variety of methods for the baseline computation has been found to yield quite different results for the end points of interest (55,77). The methods used for a given study should be prospectively defined and justified with a convincing rationale.

END POINTS FOR ANALYSES OF CENTRAL TENDENCY IN CLINICAL PHARMACOLOGY STUDIES

The primary end point of interest will be the maximum increase in the QTc interval. The optimal approach for quantifying peak QT/QTc prolongation is not a simple matter and may in fact be dependent on the pharmacokinetic and pharmacodynamic characteristics of the investigational drug in question. The recommendations in this section are intended primarily for clinical pharmacology studies in which multiple replicate ECGs have been collected over the course of a dosing interval.

When there is an obvious trend-over-time relationship, an appropriate estimate of the maximum QT/QTc prolongation effect can often be obtained at the time point at which the placebo-adjusted increase from baseline is greatest for each treatment group (ie, the mean placeboand baseline-adjusted QT/QTc value at the time point at which the one-sided 95% confidence interval upper bound is maximal). The time point at which this increase occurs may be different for each treatment group and should be specified. Time-matched placebo adjustments of this end point can be performed with comparable ease for both crossover and parallel group studies.

For such an analysis, the sponsor should be able to demonstrate that the time of maximum effect for individual subjects generally coincides with the time of maximum effect for the treatment arm, for example, by using a frequency distribution analysis of the number of subjects who experienced maximum QT/QTc prolongation at each time point. Erroneous conclusions could result if the time course of QT/QTc prolongation is subject to considerable interindividual variation, perhaps caused by variable rates of absorption, distribution, or production of active metabolites between study participants. Furthermore, conclusions based on such an analysis might be misleading if an isolated, spurious, suprathreshold spike occurs at a single point on an otherwise rather flat time-effect profile plot, with no obvious concentration relationship.

In these cases, it would be preferable to consider summary statistics based on data that were collected at time points, which can vary between subjects depending on the pharmacokinetic and pharmacodynamic characteristics of the drug in each individual. Computation of the mean change in the QT/QTc interval at the subject-specific peak plasma concentrations C^sub max^ involves determining the C^sub max^ for each subject, then identifying the change from baseline in the QT/QTc interval at the time point that coincides with or immediately follows the C^sub max^ value. In a crossover study, a time-matched, within-subject placebo adjustment would be performed. In a parallel group study, the population T^sub max^ for each active treatment group could be used to select the time point to use for placebo adjustment. If the T^sub max^ for the investigational drug is highly variable, an alternative approach to placebo adjustment, such as time-matched active group-placebo group pairs might be preferable. The baseline- and placebo-adjusted change in the QT/QTc at C^sub max^ would then be averaged for all subjects in the treatment arm.

This approach will be useful only if the maximum increase in the QT/QTc interval coincides with peak plasma concentrations. Results will be misleading if there is a substantial lag phase between the pharmacokinetic and pharmacodynamic end points because of the contribution of active metabolites, delayed distribution to myocardial tissue, or effects on ion channel processing. To support the use of this analysis, the sponsor should be able to demonstrate a strong temporal correlation between peak plasma concentration and maximum QT/QTc prolongation using superimposed mean and subject-specific time course plots for concentration of the drug and change in QT/QTc as well as hysteresis plots of concentration versus change in QT/QTc for means and individual subject data (78).

A third approach is to compute the mean of individual maximum QT/QTc interval increases (or minimum decreases for individuals who did not experience an increase). For this end point, one would examine the on-therapy placeboand baseline-adjusted QT/QTc values for all time points in each individual and select the upper limit of the range to use in computing the mean maximum increase for the treatment arm. This approach is suitable for crossover studies in which time-matched, within-subject placebo adjustment is possible. Despite the upward biasing inherent in selecting extreme values, this end point might be preferred for some drugs that have considerable interindividual variability in the time course of QT/QTc prolongation and a substantial delay between the Cmax of the parent compound and the peak QT/QTc prolongation, such that alternative end points lead to an underestimation of effect size. An inspection of subject-specific plots of the ΔQTc time profile and hysteresis plots of concentration versus QT/QTc may be useful in determining whether the observed individual maximum value shows a time and concentration relationship that is consistent with a drug effect (78).

Another end point that has been used in some ECG assessment studies is the change in the QT/QTc interval from baseline to a protocoldefined time point, representing the observed or expected time of maximum observed plasma concentration T^sub max^ for the population. This approach is discouraged because it will yield erroneous results if there is substantial interindividual variability in the pharmacokinetic T^sub max^ or a lag phase between peak plasma concentrations and maximum QT/QTc prolongation. As the T^sub max^ often varies between studies, choice of this time point on the basis of experience with previous clinical trials is inappropriate, especially when applied indiscriminantly to treatment arms receiving other drugs or doses.

Analyses of the mean time-averaged change in the QT/QTc interval are of limited value for clinical pharmacology studies in which multiple ECGs have been collected over the course of a dosing interval. Time averaging involves computing an average of all baseline- and placeboadjusted QT/QTc values for an individual over a range of time points, then determining the mean for the treatment arm using these averaged values for each individual. Time-matched placebo adjustment of this end point can be performed with comparable ease for both crossover and parallel group studies. However, the time-averaging approach ignores concentration-effect and time course relationships and underestimates the magnitude of the drug effect (79). With time-averaging, the summary statistic obtained will be critically dependent on the scheduling of the ECG recordings, such that very different effect sizes could be computed for the same treatment, depending on the time points studied. For example, the scheduling of several recordings near or subsequent to the offset of the effect would dramatically reduce the average computed.

Conceivably, two drugs might have similar maximum effects on QT/QTc prolongation but differ in terms of the rate of increase in the QT/QTc interval or the duration of time over which the increase is sustained. Such considerations might provide a partial explanation for apparent differences in proarrhythmic potential between drugs or administration routes despite similar maximum effects on QT/QTc prolongation. Therefore, in addition to comparing peak effects between treatment groups, attention should be directed to the maximum slope of the rising phase of the time-effect curve and the range of time points over which the QT/QTc interval is prolonged in relation to the placebo treatment group.

An integrated approach to quantifying the magnitude and duration of the effect is calculation of the area under the QT/QTc interval time curves (AUCs) for on-therapy versus pretherapy measurements. Successful use of the AUC is dependent on synchronization of the ECG measurement schedule for both the baseline and treatment phases. AUC values are relatively stable to random fluctuation; however, experience with this approach is limited, and interpretation of QT/QTc AUC values is complicated by the absence of well-recognized criteria for distinguishing clinically relevant absolute or delta values (79). Therefore, AUC computations in drug submissions are considered subsidiary to more established data analyses.

As the optimal approach for quantifying peak QT/QTc prolongation will vary depending on the pharmacokinetic and pharmacodynamic characteristics of the drug, the sponsor should provide results for the maximum mean increase at any postdose time point, the mean change at the subject-specific C^sub max^, and the mean of the maximum individual QT/QTc increases, together with a discussion of possible explanations for any discrepancies between the different end points. Confidence in an outcome will be increased when multiple analyses yield consistent results and when positive control agents yield signals of a magnitude that corresponds closely with expectations based on historical experience with the drug and dose in question. Complex situations can be anticipated in which the effects of the investigational drug are better described by one of the aforementioned end points, while another end point more appropriately characterizes the effects of the positive control or reference agents.

Currently, there is much discussion regarding the magnitude of QT/QTc prolongation that is considered to justify apprehensions regarding proarrhythmic potential. Various proposals have been advanced for thresholds signifying cause for concern; however, assessment of these is complicated by the fact that the observed change in the QT/QTc interval for a treatment in any given study is dependent on many factors, including, but not limited to, the following:

* The demographic characteristics of the subject population

* The electrocardiographic equipment used

* The choice of time points

* The methodology of ECG reading: (1) the lead(s) selected; (2) the conventions used for determining T-wave offset; (3) the inclusion or exclusion of large U waves in the interval measurement

* The aptitude of the ECG readers

* The method(s) used for defining the baseline QT/QTc value

* The end point(s) used for determining maximum QT/QTc prolongation

* The heart rate correction method(s)

The unavailability or inconsistent quality of data on the magnitude of peak QT/QTc prolongation for drugs of known proarrhythmic potential presents a problem when making regulatory judgments concerning small signals near the limit of study sensitivity. Even drugs that produce a relatively modest prolongation of the QT/QTc interval at therapeutic doses or low multiples thereof have been associated with events of torsade de pointes when used in patients with underlying risk factors.

For example, in a single oral dose crossover study performed in 48 healthy volunteers, mean plus or minus standard deviation changes in the QTc interval (Fridericia corrected) at the subject-specific C^sub max^ adjusted for time-matched placebo were 3.9 ±17.0 ms for 500 mg levofloxacin, 7.5 ± 15.5 ms for 1000 mg levofloxacin (1.3 times the maximum recommended single dose), and 6.0 ± 14.5 ms for 1000 mg erythromycin (twice the maximum recommended single dose) (77).

In another single oral dose crossover study performed in 48 healthy volunteers (55), the mean change in the Fridericia-corrected QTc at the subject-specific C^sub max^ appeared to be approximately 4-5 ms for 1000 mg levofloxacin (1.3 times the maximum recommended dose) and approximately 2-4 ms for 1500 mg ciprofloxacin (twice the maximum recommended single dose). Erythromycin, ciprofloxacin, and levofloxacin have all been associated with torsade de pointes during postmarketing use (80-92).

Voriconazole is an azole antifungal that has been associated with torsade de pointes both in clinical trials and during postmarketing use (93). The effects of three single oral doses of voriconazole (800, 1200, and 1600 mg) and an active comparator were investigated in healthy subjects (male and female, 18-65 years) in a randomized, single-blind, placebo-controlled, five-way crossover study (93). At 1600 mg (4 times the maximum recommended single dose), voriconazole resulted in a mean maximum increase in the Fridericia-corrected QTc of 8.23 ms (90% CI 6.01-10.45 ms).

These results would appear to suggest that even small mean increases of the QTc interval at supratherapeutic single doses are justification for concern. Additional research is encouraged to characterize the effect sizes produced by these and other compounds under conditions of steady-state dosing and optimized end point analyses.

Thresholds signifying cause for concern should be based on experience with drugs of known QT/QTc prolongation liability and borderline proarrhythmic risk in dedicated clinical pharmacology studies in which the ECGs have been collected, read, and analyzed in accordance with standardized procedures. The demonstration of reasonable consistency in results between studies performed with different subjects, investigators, readers, and equipment would be necessary for the evidence-based acceptance of thresholds.

ANALYSES OF CENTRAL TENDENCY FOR QT/QTc DATA IN PHASE II AND III CLINICALTRIALS

For therapeutic clinical trials in which ECGs were collected at periodic time points over the course of extended treatment, an analysis of change from baseline should be presented for all time points at which ECG assessments were performed. The active treatment groups should be compared with the concurrent placebo treatment group, with attention to the point estimate and the upper bound of the 90% confidence intervals at each time point. Comparisons with active control treatments are also important, especially if a placebo group is unavailable. The mean of the maximum individual on-therapy increases from baseline in each treatment group should also be computed. Presentation of an analysis of the mean time-averaged change from baseline would be acceptable only for studies in which the ECG recordings were obtained during fixed-dose treatment under steady-state conditions, with no evidence for a sustained increase or decline in the effect over the course of continued treatment. In some cases, conclusions based on data for a particular time point might be acceptable if supported by a convincing rationale (eg, single-dose parenteral use; short duration of treatment, with only one ECG assessment performed at steady state; or very long duration of treatment, such that ECGs at late time points are unlikely to be comparable to baseline ECGs because of progression of the underlying disease process). QT/QTc data that are limited to the change from baseline to final evaluation are of little value if they include ECGs collected after the last day of treatment with the study drug.

CATEGORICAL ANALYSES OF QT/QTc INTERVAL DATA

In addition to analyses of central tendency, categorical analyses should be performed to gain an impression of the proportion of study participants who meet or exceed predefined upper limit values. Outlier thresholds can be defined in terms of absolute QTc intervals or change from baseline (delta) values. Absolute interval flags are QTc values in excess of some specified threshold value. Delta flags occur when the change from baseline in the absolute QTc interval is greater than some predefined value.

The interpretation of categorical analyses of absolute interval and delta flags presents certain challenges. Treatment-emergent absolute interval QTc flags are often considered to have greater prognostic value than delta flags (94). The absolute QTc interval for a particular cardiac cycle will, however, be highly dependent on the methods used for reading the interval and performing heart rate correction. For example, a QT interval reading method that determines the interval on the basis of earliest QRS onset to latest T-wave offset in any of 12 simultaneously displayed leads would be expected to yield more absolute interval outlier signals for a given data set than a method using only one lead.

Delta values have the advantage of being much less dependent on reading method (68). Interpretation of categorical analyses for delta flags is, however, complicated by regression toward the mean. Regression toward the mean is a measurement phenomenon resulting from imperfect correlation between the baseline and postdose measurements, such that individuals having baseline QTc intervals above the mean will tend to have smaller increases from baseline than individuals with baseline QTc intervals below the mean, regardless of a treatment effect (79). This phenomenon is occasionally exploited inappropriately as apparent evidence that individuals with high baseline QTc values are less susceptible to drug-induced QTc prolongation than those with lower values. Using baseline values that are calculated from multiple measurements, rather than single readings, can reduce the effects of regression toward the mean.

For both absolute interval and delta flags, the incidence of outlier values will be dependent on the number of ECGs recorded over the treatment period and their scheduling in relationship to the time course of QTc prolongation. Comparisons with concurrent placebo and active control groups are important to place these findings in a meaningful context. Analyses should be provided for both the number and percentage of subjects with suprathreshold values (ie, number of subjects with outlier values/total number of subjects per treatment group) and the number and percentage of ECGs that exceeded threshold values (ie, number of outlier ECGs/total number of ECGs per treatment group).

Consensus within the scientific community concerning the choice of absolute QTc interval and change from baseline thresholds has remained elusive. While lower limits increase the background rate of abnormal findings, higher limits increase the risk of failing to detect a signal. Unfortunately, there is no well-recognized threshold below which a prolongation of the QTc interval is considered to be free of proarrhythmic risk. Multiple analyses using different threshold values are a reasonable approach to this controversy:

These threshold values for outlier analyses are based on experience with Bazett-corrected QTc data. Unfortunately, QTc data corrected by Bazett's formula often do not correspond closely to QTc values corrected by other formulas currently in use (76). Owing to concerns about the inaccuracy of Bazett's correction, especially at low or high heart rates, corresponding outlier thresholds for data corrected using Fridericia's formula would be highly desirable.

As noted, the incidence of abnormal values will also vary with the methods used to read the QT interval (68). Although it is difficult to declare any particular reading method superior to another, a standardized approach for regulatory purposes would facilitate the identification of outlier thresholds that could be applied consistently between studies and drug development programs.

Some debate exists whether the same threshold values should be applied to male and female subjects. Although females are known to have higher baseline QTc intervals than males (76), there is no evidence that the thresholds for proarrhythmic risk would be any greater. Indeed, female gender is a well-recognized risk factor for torsade de pointes, with women accounting for 65-75% of the drug-induced cases of this arrhythmia (36). Although use of the same thresholds for males and females would be expected to lead to a higher incidence of female outliers, this situation is likely to be informative regarding clinical risk.

Sample sizes for the dedicated QT/QTc assessment studies are computed based on the ability to detect predefined changes in the mean QTc interval. As these studies are not powered to detect outliers, the absence of extreme values should not necessarily be considered reassuring.

ANALYSES OF ECG MORPHOLOGY

Morphological abnormalities in the ECG waveform should be described and the data presented in terms of the number and percentage of subjects in each treatment group who manifested the appearance or worsening of a morphological abnormality. When a treatment-emergent effect is evident for abnormal U waves or T waves, an analysis of the number and percentage of abnormal ECGs might also be informative. When performing morphological assessments, the ECG readers should be blinded to treatment but not to patient identity because subject-specific baseline tracings serve as a point of reference when interpreting the changes occurring on treatment.

Attention should be directed to the appearance of abnormal U waves and changes in Twave morphology that might be indicative of delayed repolarization, such as double humps (biphasic or "notched" T wave), indistinct terminations (TU complex), delayed inscription (prolonged isoelectric ST segment), widening, flattening, and inversion (95). T-wave alternans (beat-to-beat variability in the amplitude, vector, or morphology of the T wave) is considered a harbinger of ventricular arrhythmias.

INTEGRATED ANALYSES

Integrated analyses can provide useful information on the adequacy of the ECG safety database in terms of the total number of patients receiving ECG evaluations (eg, per dose group, indication, subpopulation, etc.) as well as overall estimates of effect size and the incidence of outlier values. Analyses of pooled ECGs from several clinical trials are appropriate, provided that the ECG assessment procedures were of comparable rigor. Standardization of the ECG collection schedule (eg, number and frequency of visits, timing of ECG recordings in relation to dosing) for similar studies within a clinical trial program will facilitate pooled analyses.

The clinical trials used in the generation of such analyses should be clearly identified and their inclusion justified. The data from certain trials or treatment groups may be inappropriate for pooling if the study conditions under which they were obtained were not representative of the proposed clinical use. For example, if the pooling results in the inclusion of data from many patients receiving subtherapeutic doses of the drug, the magnitude of QT/QTc prolongation and the incidence of outliers at the recommended therapeutic doses would be underestimated. To avoid variability introduced by investigators operating from different regions and centers, the ECGs used for the integrated analysis should be assessed by a central laboratory at which a uniform methodology for reading and interpretation can be enforced.

Subgroup analyses of the pooled QT/QTc data from the phase II and III clinical trials are desirable for drugs that delay ventricular repolarization. The availability of subgroup analyses for gender, age (eg, younger than 18 years, 65 years and older), cardiac comorbidities, hepatic impairment, renal impairment, and other special patient populations is recommended. Such subgroup analyses should be provided for both analyses of central tendency and categorical analyses. For many special populations, such as persons with renal or hepatic impairment, the number of patients in phase II and III trials will often be too small to permit a meaningful subgroup analysis. The inclusion of intensive ECG monitoring in the pharmacokinetic studies performed in these special populations will often provide the best opportunity to examine drugdisease interactions affecting QT/QTc prolongation.

Scatter plots of temporally paired QT/QTc and serum potassium values might be a useful approach for examining correlations between these safety parameters.

ADVERSE EVENTS

The ICH E14 guideline lists some adverse events that are suggestive or indicative of proarrhythmia secondary to QT/QTc prolongation (eg, torsade de pointes, sudden death, ventricular tachycardia, ventricular fibrillation, ventricular flutter, syncope, and seizures).

MedDRA® the Medical Dictionary for Regulatory Activities terminology is the international medical terminology developed under the auspices of the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH). Examples of MedDRA preferred terms for adverse events that are of interest in this regard are listed in Table 2.

An excess in the incidence of such events for the new drug when compared with the placebo or active control groups may be justification for concern. Detailed patient narratives should be provided for all serious cardiac adverse events. In assessing the possible causal relationship of drug-induced QT/QTc prolongation, attention should be directed to considerations such as temporal relationship, dose dependency, evidence of positive dechallenge or rechallenge, corroborative nonclinical findings, and ECG results, preferably collected at the time of the event. As the QT/QTc interval is subject to considerable fluctuation, a possible role for QT/ QTc prolongation cannot be dismissed on the basis of normal on-therapy ECG recordings performed prior or subsequent to the adverse event. Potential relationships between the occurrence of the event and patient age, gender, preexisting cardiac disease, electrolyte abnormalities, concomitant medications, and other risk factors should be explored.

Consideration should be given to the influence of the eligibility criteria on the risk of QT/QTc prolongation and associated adverse events in the study population (eg, exclusion of patients with cardiac comorbidities). Ideally, the phase II and III studies should include an adequate representation of female and elderly patients.

Standard criteria should be specified in clinical trial protocols for defining what constitutes an adverse event with respect to QT/QTc prolongation and changes in T wave morphology. If a subject experiences symptoms or has ECG findings suggestive of an arrhythmia during a clinical trial, an urgent specialist assessment should be arranged. Immediate discontinuation of the suspect drug will often be appropriate.

As torsade de pointes and sudden cardiac death are expected to be quite rare events with most non-antiarrhythmic drugs that cause QT/QTc prolongation, the failure to observe these events over the course of a typical clinical trial program is not a sufficient basis for dismissing possible proarrhythmic risks when these are suspected on the grounds of ECG data. Owing to their rarity, serious ventricular arrhythmias with a QT/QTc-prolonging drug are often reported only after large populations of patients have received the agent in postmarketing settings.

PHARMACOGENETIC CONSIDERATIONS

Many forms of congenital LQTS are associated with mutations in the genes that encode cardiac ion channels or associated regulatory proteins. As these disorders are thought to be risk factors for an exaggerated response to QT/QTc-prolonging drugs, genotyping should be considered for subjects who experience marked QT/ QTc prolongation or symptoms suggestive of proarrhythmia while receiving treatment with an investigational drug. Table 3 summarizes the genetic, biochemical, and functional basis of the various forms of congenital LQTS that have been identified to date (96-98).

Because of incomplete penetrance, not all carriers of mutated ion channel genes will manifest QT/QTc prolongation in screening ECG evaluations (99). In addition to mutations, polymorphisms affecting cardiac ion channel proteins can result in increased susceptibility to drug-induced QT/QTc prolongation (100,101). Routine collection of deoxyribonucleic acid (DNA) samples from consenting subjects will allow for genotyping of candidate genes in individuals who are subsequently found to exhibit marked QT/QTc prolongation, morphological changes, or symptoms suggestive of arrhythmia while receiving an investigational drug.

REGULATORY IMPLICATIONS

Regulatory decisions regarding approval and prescribing information should be based on a careful assessment of relevant data from all stages of drug development, with appropriate attention to evidence of dose dependency, concentration relationship, and trend over time; central tendency analyses of effect size; categorical analyses of outlier values; morphological abnormalities; discontinuations and dosage reductions caused by QT/QTc prolongation; and pre- or postmarketing adverse events suggestive of proarrhythmia.

Substantial prolongation of the QT/QTc interval, with or without documented arrhythmias, can be the basis for nonapproval of a drug or discontinuation of its clinical development. Failure to perform an adequate clinical assessment of the QT/QTc prolongation liability of a drug may likewise be justification for delaying or denying marketing authorization.

The risk-benefit assessment of a QT/QTc-prolonging drug will take into account the morbidity and mortality associated with the untreated disease or condition, the clinical significance of the beneficial effects, and the magnitude of the QT/QTc interval prolongation. The demonstration of therapeutic benefits in patients who are refractory to, intolerant of, or not candidates for available therapy might justify the approval of a QT/QTc-prolonging drug if its indication was limited to use in such patients.

Considerations that will have a negative impact on the risk-benefit assessment include the availability of therapeutic alternatives that are otherwise similar in terms of safety and efficacy but lack QT/QTc prolongation effects. If QT/ QTc prolongation is a feature shared by other drugs of the therapeutic class in question, the risk-benefit assessment will involve a comparison of the effects observed with the new drug to those of its class members in concurrent active control groups.

Concerns would be intensified by a steep dose- or concentration-effect relationship that would make dosing errors or normal interindividual variability particularly dangerous. Drugs with primary metabolic pathways involving enzymes that are subject to genetic polymorphisms (eg, CYP2D6, CYP2C19) or inhibition by many drugs (eg, CYP3A4) would be of particular concern in this regard because of the risk of markedly elevated plasma levels in poor metabolizers or in the presence of interacting xenobiotics. A susceptibility to drug-drug interactions at the level of transporter proteins would also have a negative impact on the risk-benefit assessment.

The utility and feasibility of risk management options are important final considerations in rendering a decision about the approvability of a drug that prolongs the QT/QTc interval.

The recent regulatory guideline initiatives have stimulated considerable progress in the design of QT/QTc assessment studies, the acquisition and reading of ECGs, and the statistical analysis of QT/QTc data. The information derived from these studies will have an important role in guiding decision making by pharmaceutical companies, regulatory authorities, and prescribing physicians and fostering improved cardiac safety for new drugs.

*The views expressed in this article are lhose of the author and do not necessarilv reflect the opinions of the International Conference on Harmonisation (ICH) Expert Working Groups or Steering Committee, Health Canada, other regulatory authorities, or their advisory bodies.

REFERENCES

1. Bednar MM, Harrigan EP, Anziano RJ, Camm AJ, Ruskin JN. The QT interval. Prog Cardiovasc Dis. 2001;43(5 suppl 1):1-45.

2. Haverkamp W, Monnig G, Schulze-Bahr E, Haverkamp F, Breithardt G. Physician-induced torsade de pointes-therapeutic implications. Cardiovasc Drugs Ther. 2002:16:101-109.

3. Yap YG, Camm AJ. Drug induced QT prolongation and torsade de pointes. Heart. 2003;89: 1363-1372.

4. Roden DM. Drug-induced prolongation of the QT interval. N Engl J Med. 2004:350:1013-1022.

5. Shah RR. Drug-induced QT interval prolongation: regulatory perspectives and drug development. Ann Med. 2004;36(suppl 1):47-52.

6. The University of Arizona Center for Education and Research on Therapeutics. Drugs that Prolong the Qt Interval and/or Induce Torsades de Pointes Ventricular Arrhythmia. Available at: http://www.torsades.org/medical-pros/drug-lists/ drug-lists.htm. Accessed August 28, 2005.

7. Redfern WS, Carlsson L, Davis AS, et al. Relationships between preclinical cardiac electrophysiology, clinical QT interval prolongation and torsade de pointes for a broad range of drugs: evidence for a provisional safety margin in drug development. Cardiovasc Res. 2003:58:32-45.

8. De Bruin ML, Pettersson M, Meyboom RHB, Hoes AW, Leufkens HGM. Anti-HERG activity and the risk of drug-induced arrhythmias and sudden death. Eur Heart J. 2005:26:590-597.

9. Schlepper M. Cardiodepressive effects of antiarrhythmic drugs. Eur Heart J. 1989;10(suppl E): 73-80.

10. Nau C, Seaver M, Wang S-Y, Wang GK. Block of human heart hHl sodium channels by amitriptyline. J Pharmacol Exp Ther. 2000:292:1015-1023.

11. Kuryshev YA, Brown AM, Wang L, Benedict CR, Rampe D. Interactions of the 5-hydroxytryptamine 3 antagonist class of antiemetic drugs with human cardiac ion channels. J Pharmacol Exp Ther. 2000:295:614-620.

12. Gamouras GA, Monir G, Plunkitt K, Gursoy S, Dreifus LS. Cocaine abuse: repolarization abnormalities and ventricular arrhythmias. Am J Med Sci. 2000:320:9-12.

13. Bauman JL, DiDomenico RJ. Cocaine-induced channelopathies: emerging evidence on the multiple mechanisms of sudden death. J Cardiovasc Pharmacol Ther. 2002:7:195-202.

14. Gintant GA. Azimilide causes reverse rate-dependent block while reducing both components of delayed-rectifier current in canine ventricular myocytes. J Cardiovasc Pharmacol. 1998:31: 945-953.

15. Chouabe C, Drici MD, Romey G, Barhanin J. Effects of calcium channel blockers on cloned cardiac K+ channels I^sub Kr^ and I^sub Ks^. Therapie. 2000;55: 195-202.

16. Kang J, Chen XL, Wang L, Rampe D. Interactions of the antimalarial drug mefloquine with the human cardiac potassium channels KvLQT 1/minK and HERG. J Pharmacol Exp Ther. 2001:299:290-296.

17. Yildirim H, Adanir T, Atay A, Katircioglu K, Savaci S. The effects of sevoflurane, isoflurane and desflurane on QT interval of the ECG. Eur J Anaesthesiol. 2004:21:566-570.

18. Suzuki A, Bosnjak ZJ, Kwok WM. The effects of isoflurane on the cardiac slowly activating delayed-rectifier potassium channel in guinea pig ventricular myocytes. Anesth Analg. 2003;96: 1308-1315.

19. Shibata S, Ono K, lijima T. Sevoflurane inhibition of the slowly activating delayed rectifier K+ current in guinea pig ventricular cells. J Pharmacol Sci. 2004:95:363-373.

20. Kuhlkamp V, Mewis C, Bosch R, Seipel L. Delayed sodium channel inactivation mimics long QT syndrome 3. J Cardiovasc Pharmacol. 2003;42:113-117.

21. Picker E, Kuryshev YA, Dennis AT, et al.. Mechanisms of arsenic-induced prolongation of cardiac repolarization. Mol Pharmacol. 2004;66:33-44.

22. Kuryshev YA, Ficker E, Wang L, et al. Pentamidine-induced long QT syndrome and block of hERG trafficking. J Pharmacol Exp Ther. 2005; 312:316-323.

23. Lee S, Harris ND, Robinson RT, Yeoh L, Macdonald IA, Heller SR. Effects of adrenaline and potassium on QTc interval and QT dispersion in man. Eur J Clin Invest. 2003:33:93-98.

24. Totterman KJ, Huhti L, Sutinen E, et al. Tolerability to high doses of formoterol and terbutaline via Turbuhaler for 3 days in stable asthmatic patients. EurRespirJ. 1998:12:573-579.

25. Kuusela TA, Jartti TT, Tahvanainen KU, Kaila TJ. Prolongation of QT interval by terbutaline in healthy subjects. J Cardiovasc Pharmacol. 2005;45: 175-181.

26. Karle CA, Zitron E, Zhang W, Kathofer S, Schoels W, Kiehn J. Rapid component I(Kr) of the guineapig cardiac delayed rectifier K(+) current is inhibited by β(1)-adrenoreceptor activation, via cAMP/protein kinase ?-dependent pathways. Cardiovasc Res. 2002 1:53:355-362.

27. McBride BF, Karapanos AK, Krudysz A, Kluger J, Coleman CI, White CM. Electrocardiographic and hemodynamic effects of a multicomponent dietary supplement containing ephedra and caffeine: a randomized controlled trial. JAMA. 2004; 291:216-221.

28. Nasir JM, Durning SJ, Ferguson M, Barold HS, Haigney MC. Exercise-induced syncope associated with QT prolongation and ephedra-free Xenadrine. Mayo Clin Proc. 2004;79:1059-1062.

29. Singh BN, Hollenberg NK, Poole-Wilson PA, Robertson JI. Diuretic-induced potassium and magnesium deficiency: relation to drug-induced QT prolongation, cardiac arrhythmias and sudden death. J Hypertens. 1992;10:301-316.

30. Charbit B, Funck-Brentano C, Samain E, JannierGuillou V, Albaladejo P, Marty J. QT interval prolongation after oxytocin bolus during surgical induced abortion. Clin Pharmacol Ther. 2004; 76: 359-364.

31. Eden E, Teirstein A, Wiener I. Ventricular arrhythmia induced by vasopressin: torsade de pointes related to vasopressin-induced bradycardia. Mi SinaiJMed. 1983;50:49-51.

32. Mauro VF, Bingle JF, Ginn SM, Jafri FM. Torsade de pointes in a patient receiving intravenous vasopressin. Grit CareMed. 1988:16:200-201.

33. Eckert B, Agardh CD. Hypoglycaemia leads to an increased QT interval in normal men. Clin Physiol. 1998:18:570-575.

34. Robinson RT, Harris ND, Ireland RH, Lee S, Newman C, Heller SR. Mechanisms of abnormal cardiac repolarization during insulin-induced hypoglycemia. Diabetes. 2003:52:1469-1474.

35. FDA review for Plenaxis (abarelix) injectable suspension. Available at: www.fda.gov/cder/foi/nda/ 2003/21-320_Plenaxis.htm. Accessed March 26, 2005.

36. Abi-Gerges N, Philp K, Pollard C, Wakefield 1, Hammond TG, Valentin JP. Sex differences in ventricular repolarization: from cardiac electrophysiology to Torsades de Pointes. Fundam Clin Pharmacol. 2004:18:139-151.

37. Carnethon MR, Anthony MS, Cascio WE, et al. A prospective evaluation of the risk of QT prolongation with hormone replacement therapy: the atherosclerosis risk in communities study. Ann Epidemiol. 2003:13:530-536.

38. Kadish AH, Greenland P, Limacher MC, Frishman WH, Daugherty SA, Schwartz JB. Estrogen and progestin use and the QT interval in postmenopausal women. Ann Noninvasive Electrocardiol. 2004:9:366-374.

39. Miyamoto S, Zhu BM, Aye NN, Hashimoto K. Slowing Na+ channel inactivation prolongs QT interval and aggravates adrenaline-induced arrhythmias. Jpn J Pharmacol. 2001 ;86: 114-119.

40. Carbone E, Wanke E, Prestipino G, Possani LD, Maelicke A. Selective blockage of voltage-dependent K+ channels by a novel scorpion toxin. Nature (London). 1982:296:90-91.

41. Gurrola GB, Rosati B, Rocchetti M, et al. A toxin to nervous, cardiac, and endocrine ERG K+ channels isolated from Centruroides HOJCIUS scorpion venom. FASEBJ. 1999:13:953-962.

42. Guatteo E, Faravelli L, Verotta L, Pelizzoni F, Rogers CB, Wanke K. A novel K+ channel blocker isolated from "hiccup nut" toxin. NeuroReport. 1996:7:2575-2579.

43. Fenichel RR, Malik M, Antzelevitch C, et al. Druginduced torsades de pointes and implications for drug development. J Cordiovosc Electrophysiol, 2004:15:475-495.

44. Herron W, Towers C, Templeton A. Experiences in method development for the analysis of in vitro study solutions for content. J Pharmacol Toxicol Methods. 2004:49:211-216.

45. Kirsch GE, Trepakova ES, Brimecombe JC, et al. Variability in the measurement of hERG potassium channel inhibition: effects of temperature and stimulus pattern. J Pharmacol Toxicol Methods. 2004:50:93-101.

46. Witchel HJ, Milnes JT, Mitcheson JS, Hancox JC. Troubleshooting problems with in vitro screening of drugs for QT interval prolongation using HERG K+ channels expressed in mammalian cell lines and Xenopus oocytes. J Pharmacol Toxicol Methods. 2002;48:65-80.

47. Titier K, Canal M, Deridet E, et al. Determination of myocardium to plasma concentration ratios of five antipsychotic drugs: comparison with their ability to induce arrhythmia and sudden death in clinical practice. Toxicol Appl Pharmacol. 2004; 199:52-60.

48. Pratt C, Brown AM, Rampe D, et al. Cardiovascular safety of fexofenadine HCl. Clin Exp Allergy. 1999;29(suppl 3):212-216.

49. Sanguinetti MC, Mitcheson JS. Predicting drughERG channel interactions that cause acquired long QT syndrome. Trends Pharmacol Sd. 2005; 26:119-124.

50. Oberg KC, Bauman JL. QT interval prolongation and torsades de pointes due to erythromycin lactobionate. Phamacotherapy. 1995;15:687-692.

51. Rezkalla MA, Pochop C. Erythromycin induced torsades de pointes: case report and review of the literature. S D J Med. 1994:47:161-164.

52. Hassaballa HA, Balk RA. Torsade de pointes associated with the administration of intravenous haloperidol:a review of the literature and practical guidelines for use. Expert Opin Drug Saf. 2003:2:543-547.

53. Lawrence KR, Nasraway SA. Conduction disturbances associated with administration of butyrophenone antipsychotics in the critically ill: a review of the literature. Pharmacotherapy. 1997; 17:531-537.

54. Malik M, Hnatkova K, Batchvarov V, Gang Y, Smetana P, Camm AJ. Sample size, power calculations, and their implications for the cost of thorough studies of drug induced QT interval prolongation. PACE. December 27, 2004:1659-1669.

55. Noel GJ, Natarajan J, Chien S, Hunt TL, Goodman DB, Abels R. Effects of three fluoroquinolones on QT interval in healthy adults after single doses. CKn Pharmacol Ther. 2003:73:292-303.

56. Harrigan EP, Miceli JJ, Anziano R, et al. A randomized evaluation of the effects of six antipsychotic agents on QTc, in the absence and presence of metabolic inhibition. J Clin Psychopharmacol. 2004:24:62-69.

57. Morganroth J, Ilson BE, Shaddinger BC, et al. Evaluation of vardenafil and sildenafil on cardiac repolarization. Am } CardioL 2004;93:1378-1383, A6.

58. Zhou Z, Vorperian VR, Gong Q, Zhang S, January CT. Block of HERG potassium channels by the antihistamine astemizole and its metabolites desmethylastemizole and norastemizole. J Cardiovasc Electrophysiol. 1999:10:836-843.

59. Barbey JT, Pezzullo JC, Soignet SL. Effect of arsenic trioxide on QT interval in patients with advanced malignancies. J Clin Oncol. 2003:21:3609-3615.

60. Cooney MM, Radivoyevitch T, Dowlati A, et al. Cardiovascular safety profile of combretastatin a4 phosphate in a single-dose phase I study in patients with advanced cancer. Clin Cancer Res. 2004:10(1 pt 1):96-100.

61. Varterasian M, Fingert H, Agin M, Meyer M. Consideration of QT/QTc interval data in a phase I study in patients with advanced cancer. Ciin Cancer Res. 2004:10:5967-5968.

62. Krantz MJ, Kutinsky IB, Robertson AD, Mehler PS. Dose-related effects of methadone on QT prolongation in a series of patients with torsade de pointes. Pharmacotherapy. 2003 ;23: 802-805.

63. Bedigian MP. Use of ECGs in support of cardiac safety in phase II and III clinical trials. In: Morganroth J, Gussak I, eds. Cardiac Safety ofNoncardiac Drugs: Practical Guidelines for Clinical Research and Drug Development. Totowa, NJ: Humana Press; 2005:223-227.

64. Mortara JL. ECG acquisition and signal processing. In: Morganroth J, Gussak I, eds. Cardiac Safety ofNoncardiac Drugs: Practical Guidelines for Clinical Research and Drug Development. Totowa, NJ: Humana Press; 2005:131-145.

65. Brown BD. The PDA's digital ECG initiative and its impact on clinical trials. In: Morganroth J, Gussak I, eds. Cardiac Safety ofNoncardiac Drugs: Practical Guidelines for Clinical Research and Drug Development. Totowa, NJ: Humana Press; 2005: 301-327.

66. Lau CP, Freedman AR, Fleming S, et al. Hysteresis of the ventricular paced QT interval in response to abrupt changes in pacing rate. Cardiovasc Res. 1988:22:67.

67. Malik M. Errors and misconceptions in ECG measurement used for the detection of drug induced QT interval prolongation. J Electrocardiol. 2004;37(suppl):25-33.

68. Azie NE, Adams G, Darpo B, et al. Comparing methods of measurement for detecting drug-induced changes in the QT interval: implications for thoroughly conducted ECG studies. Ann Noninvasive Electrocardiol. 2004;9:166-174.

69. Malik M, Camm AJ. Evaluation of drug-induced QT interval prolongation. Implications for drug approval and labelling. Drug Saf. 2001 ;24: 323-351.

70. Noel GJ, Goodman DB, Chien S, Solanki B, Padmanabhan M, Natarajan J. Measuring the effects of supratherapeutic doses of levofloxacin on healthy volunteers using four methods of QT correction and periodic and continuous ECG recordings. J Clin Pharmacol. 2004;44:464-473.

71. Malik M, Farbom P, Batchvarov V, Hnatkova K, Camm AJ. Relation between QT and RR intervals is highly individual among healthy subjects: implications for heart rate correction of the QT interval. Heart. 2002;87:220-228.

72. Couderc J-P, Xiaojuan X, Zareba W, Moss AJ. Assessment of the stability of the individual-based correction of QT interval for heart rate. Ann Noninvasive Electrocardiol. 2005;10:25-34.

73. Badilini F, Maison-Blanche P. Holter monitoring for QT. The RR bin method in depth. In: Morganroth J, Gussak I, eds. Cardiac Safety ofNoncardiac Drugs: Practical Guidelines for Clinical Research and Drug Development. Totowa, NJ: Humana Press; 2005:167-185.

74. FDA review for Geodon (Ziprasidone HCl capsules). Available at: http://www.fda.gov/cder/foi/ nda/2001/20-825_Geodon.htm. Accessed March 28, 2005.

75. FDA review for Uroxatral (alfuzosin HCl tablets). Available at: http://www.fda.gov/cder/foi/ nda/2003/021287_uroxatral_toc.htm. Accessed March 28, 2005.

76. Luo S, Michler K, Johnston P, Macfarlane PW. A comparison of commonly used QT correction formulae: the effect of heart rate on the QTc of normal ECGs. / Electrocardiol. 2004;37(suppl):81-90.

77. FDA review for Avelox IV in sodium chloride 0.8% in plastic container. Available at: http://www .fda.gov/cder/foi/nda/2001/21277_Avelox.htm. Accessed on March 28, 2005.

78. Hollister AS, Montague TH. Statistical analysis plans for ECG data. Controlling the intrinsic and extrinsic variability in QT data. In: Morganroth J, Gussak I, eds. Cardiac Safety of Noncardiac Drugs: Practical Guidelines for Clinical Research and Drug Development. Totowa, NJ: Humana Press; 2005: 239-257.

79. Bonate PL, Russell T. Assessment of QTc prolongation for non-cardiac-related drugs from a drug development perspective. J Clin Pharmacol. 1999:39:349-358.

80. Shaffer D, Singer S, Korvick J, Honig P. Concomitant risk factors in reports of torsades de pointes associated with macrolide use: review of the United States Food and Drug Administration Adverse Event Reporting System. CHn Infect Dis. 2002:35:197-200.

81. Ray WA, Murray KT, Meredith S, Narasimhulu SS, Hall K, Stein CM. Oral erythromycin and the risk of sudden death from cardiac causes. N Engl J Med. 2004:351:1089-1096.

82. Koh TW. Risk of torsades de pointes from oral erythromycin with concomitant carbimazole (methimazole) administration. Pacing Clin Electrophysiol. 2001:24:1575-1576.

83. Lin JC, Quasny HA. QT prolongation and development of torsades de pointes with the concomitant administration of oral erythromycin base and quinidine. Pharmacotherapy. 1997; 17: 626-630.

84. Cranberry MC, Gardner SF. Erythromycin monotherapy associated with torsade de pointes. Ann Pharmacother. 1996:30:77-78.

85. Berger RE. Rates of torsades de pointes associated with ciprofloxacin, ofloxacin, levofloxacin, gatifloxacin, and moxifloxacin. J Urol. 2005:174:165.

86. Frothingham R. Rates of torsades de pointes associated with ciprofloxacin, ofloxacin, levofloxacin, gatifloxacin, and moxifloxacin. Pharmacotherapy. 2001 ;21:1468-1472.

87. Owens RC Jr, Ambrose PG. Torsades de pointes associated with fluoroquinolones. Pharmacotherapy. 2002:22:663-668; discussion 668-672.

88. Katritsis D, Camm AJ. Quinolones: cardioprotective or cardiotoxic. Pacing CHn Electrophysiol. 2003:26:2317-2320.

89. Iannini PB. Cardiotoxicity of macrolides, ketolides and fluoroquinolones that prolong the QTc interval. Expert Opin Drug Saf. 2002;1: 121-128.

90. Gandhi PJ, Menezes PA, Vu HT, Rivera AL, Ramaswamy K. Fluconazole- and levofloxacin-induced torsades de pointes in an intensive care unit patient. Am J Health Syst Pharm. 2003; 60:2479-2483.

91. Prabhakar M, Krahn AD. Ciprofloxacin-induced acquired long QT syndrome. Heart Rhythm. 2004:1:624-626.

92. Daya SK, Gowda RM, Khan IA. A ciprofloxacinand hypocalcemia-induced torsade de pointes triggered by hemodialysis. Am J Ther. 2004; 11: 77-79.

93. Vfend Product Monograph. Available at: http://www.pfizer.ca/english/our%20products/ prescription%20pharmaceuticals. Accessed on June 22, 2005.

94. Shah RR. Interpretation of clinical ECG data. Understanding the risk from non-antiarrhythmic drugs. In: Morganroth J, Gussak I, eds. Cardiac Safety ofNoncardiac Drugs: Practical Guidelines for Clinical Research and Drug Development. Totowa, NJ: Humana Press; 2005:259-298.

95. Ackerman MJ. Ion channel diseases of the heart. Mayo Clin Proc. 1998:73:250-269.

96. Ackerman MJ. Molecular basis of congenital and acquired long QT syndromes. J Electrocardiol. 2004;37(suppl):1-6.

97. Shah RR. Pharmacogenetic aspects of drug-induced torsade de pointes: potential tool for improving clinical drug development and prescribing. Drug Saf. 2004;27:145-172.

98. Moss AJ, Kass PS. Long QT syndrome: from channels to cardiac arrhythmias. J Clin Invest. 2005:115:2018-2024.

99. Priori SG, Napolitano C, Schwartz PJ. Low penetrance in the long-QT syndrome: clinical impact. Circulation. 1999:99:529-533.

100. Anson BD, Ackerman MJ, Tester DJ, et al. Molecular and functional characterization of common polymorphisms in HERG (KCNH2) potassium channels. Am ) Physiol Heart Circ Physiol. 2004;286:H2434-H2441.

101. Sesti F, Abbott GW, Wei J, et al. A common polymorphism associated with antibiotic-induced cardiac arrhythmia. Proc Natl Acad Sci USA. 2000:97:10,613-10,618.

Colette Strnadova, PhD

Therapeutic Products Directorate. Health Canada, Ottawa, Ontario, Canada

Correspondence Address

Colette Stmadova, PhD, Therapeutic Products Directorate, Health Canada, 2nd Floor, Holland Cross. Tower B. 1600 Scott Street, A.L 3102C3, Ottawa. Ontario. Canada, K1A OK9 (email: colelte_stmad@hc-sc.gc.ca)

Copyright Drug Information Association 2005
Provided by ProQuest Information and Learning Company. All rights Reserved