Many noncardiovascular drugs have been withdrawn or had their labeling changed due to risks of sudden cardiac death linked with QTc prolongation.1,2 In most cases, drug-induced QT/QTc prolongation appears to result from a blockade of the delayed rectifier potassium (IKr) ion channel, which is involved in cardiac repolarization and coded by the human ether-a-go-go-related gene (hERG).3 Because of the potential clinical and regulatory implications of drug-induced QT prolongation, the rigor for screening new drug candidates for this effect throughout drug development has increased. In particular, multiple regulatory initiatives have been undertaken over the past several years, resulting in consensus guidelines for both preclinical evaluations, including assays for hERG inhibitory activity,4 and clinical evaluations, including “thorough QT/QTc studies.”5
Through development of the International Conference on Harmonization (ICH) E14 guideline, an initial, consistent approach to the design of the thorough QT/QTc study was established. In particular, per the guidance, the thorough QT study is usually performed in a double-blind randomized fashion in healthy volunteers, uses doses that produce concentrations similar to those seen under conditions producing maximum exposure, including metabolic inhibition, is placebo-controlled, and has a positive control. The guidance also notes that subgroups of individuals with risk factors for QT prolongation, such as women, individuals with impaired clearance or metabolizing capacity, and individuals >65 years of age, may be of particular interest.
Duloxetine is a serotonin (5-HT) and norepinephrine (NE) reuptake inhibitor (SNRI) with no significant affinity for other receptors.6-8 It is approved as an effective treatment for depression in the US and EU,9-15 the management of diabetic neuropathic pain in the US and EU,16-18 and the treatment of stress urinary incontinence in EU.19-21
Duloxetine is well-absorbed, reaching steady-state plasma concentrations (mean ± SD) of 48.5 ± 28.9 ng/mL during once-daily (QD) dosing of 60 mg and 72.6 ± 43.2 ng/mL during twice-daily (BID) dosing of 40 mg. On the basis of duloxetine's elimination half-life of 12.5 hours,22 steady-state concentrations are reached within 3 days of repeated dosing. Duloxetine is metabolized primarily by CYP1A2 with minor contribution from CYP2D6 to 2 major metabolites-the glucuronide conjugate of 4-hydroxy duloxetine and the sulfate conjugate of 5-hydroxy-6-methoxy duloxetine. Neither metabolite is pharmacologically active.23 Like the parent compound, peak plasma concentrations of these 2 metabolites are typically reached about 6 hours after dosing in fasted subjects.24
During clinical studies that evaluated cardiac safety, duloxetine at doses up to 120 mg/day did not affect cardiac repolarization as measured by QTc interval.25,26 In addition, in vitro ion channel studies showed that duloxetine had no effect on any of the ion currents tested at concentrations below 18 μM, except for hERG, where a significant blockade occurred at 1.8 μM with an IC50 value of 5.5 μM. At the concentrations measured in humans during duloxetine treatment at dosages up to 60 mg BID, duloxetine had no effect on any of the human cardiac ion channels tested (Lilly Research Laboratories, data on file). These results suggested that at therapeutic plasma concentrations, duloxetine is very unlikely to produce arrhythmias by blocking ion channels. Toxicology data in rats and dogs also predicted no clinical concern regarding heart rate, blood pressure, or cardiac conduction in humans (Lilly Research Laboratories, data on file).
However, in order to satisfy regulatory criteria during the development of duloxetine, a clinically thorough QT/QTc study was conducted to evaluate QT/QTc effects with the drug at supratherapeutic doses. Specifically, the preliminary studies examined (1) the duloxetine exposures reached at therapeutic dose levels under conditions of maximal metabolic inhibition in CYP2D6 poor metabolizers treated with the potent CYP1A2 inhibitor fluvoxamine,27 and (2) the exposures and tolerability of ascending supratherapeutic dose levels.28 These studies indicated that a duloxetine dosage of 200 mg BID would be expected to produce steady-state exposures comparable to those reached under maximal metabolic inhibition.27,28
The objective of this study was to confirm that the mean QTc interval occurring with a supratherapeutic dose of duloxetine did not differ from the mean QTc interval for placebo, as evidenced by the upper bound of the 2-sided 90% confidence interval of the mean difference of less than 10 msec. Moxifloxacin, a fluoroquinolone antibiotic known to prolong the QTc interval,29 was used as the positive control to confirm sufficient assay sensitivity.
MATERIALS AND METHODS
This was a multicenter, double-blind, randomized, crossover trial of duloxetine and placebo conducted in the United States (Figure 1A). The subjects were randomly assigned in a 1:1:1:1 ratio to 1 of 4 treatment groups. After a baseline day, subjects in groups 1 and 3 received duloxetine for 20 days followed by placebo for 22 days; subjects in groups 2 and 4 received placebo and then duloxetine. In all 4 groups, a washout period of up to 14 days followed completion of the placebo/duloxetine study period. Subjects were randomly assigned to receive a single oral dose of 400 mg of moxifloxacin either before or after the placebo/duloxetine study period followed by a 3 day washout period. The protocol was approved by Institutional Review Boards at each study site, and the study was conducted in accordance with the Declaration of Helsinki.
The study population consisted of healthy female volunteers between the ages of 18 and 75 years. Subjects with clinically significant laboratory abnormalities were excluded. All subjects had normal electrocardiograms (Bazett's QTc intervals < 450 msec), blood pressure, and heart rate (supine and standing) as determined by the investigator. Every subject gave written informed consent.
Duloxetine was administered to a maximum dose level of 200 mg BID. To safely achieve this dose level, subjects progressed through a dose-escalation procedure in which duloxetine was administered in 4 day consecutive intervals of 60 mg BID (level 1), 120 mg BID (level 2), 160 mg BID (level 3), and 200 mg BID (level 4) (Figure 1B). This stepwise dose-escalation allowed the investigators to monitor individual safety as the dose progressed to the maximum level. Subjects who failed to tolerate treatment during the first 2 dosing levels were discontinued from the study. During the first 20 day dosing period, subjects had a 4 day taper period while receiving duloxetine 120 mg BID for 2 days and then 60 mg BID for 2 days, or the corresponding placebo for 4 days. During the 22 day dosing period, subjects had a 6 day taper period while receiving duloxetine 160 mg BID for 2 days, 120 mg BID for 2 days, and 60 mg BID for 2 days or the corresponding placebo for 6 days.
Evaluation of QTc
Standard digital electrocardiograms (ECG) were collected using the MAC 5000, MAC VU Cardiographs, and Muse CV Cardiology System (GE Medical Systems Milwaukee, WI)30 and transmitted electronically to the ECG core lab for over read. At the core lab, the Marquette12 SL ECG Analysis Program creates a global superimposed median beat for measurement of the ECG intervals (PR, QRS, and QT) from simultaneously sampled data from all 12 leads of the 10-second recording. The program identifies the QRS onset and the T-wave offset, and the analysis employs magnification and allows for on-screen computer-assisted evaluation and editing of QT/QTc intervals. To minimize interindividual variability in the data, digital ECGs were over-read by a single core lab cardiologist trained in a standardized approach for interpreting the ECG data and blinded to subject demographic data as well as treatment assignment.
To account for intrinsic variability of QTc measurements, 4 replicate resting 12-lead ECG measurements were collected at each of 4 time points (2, 6, 10, and 12 hours) on the baseline day, on the fourth day of dosing of duloxetine or placebo at the 2 highest dose levels (160 mg BID and 200 mg BID), and at 2 and 6 hours after dosing with moxifloxacin. These resting ECGs were obtained at specific sampling times spaced throughout a dosing interval and included expected maximum duloxetine concentrations at 6-10 hours. At each time point, sequential ECGs were obtained consecutively until the investigator determined that 4 of these recordings were suitable for measurement of QT intervals. An additional fifth ECG was then collected for immediate safety evaluation.
The purpose of the pharmacokinetic analysis was to confirm that exposures to duloxetine and its 2 major metabolites were sufficiently high compared with exposures in patients taking duloxetine under prescribed conditions. Blood samples for pharmacokinetic analysis of plasma duloxetine and its 2 major metabolites were collected from each subject on the fourth day of dosing at dose levels 3 and 4 at time points that coincided with ECG assessments (2, 6, 10, and 12 hours). Individual concentration-time profiles of duloxetine over a 12 hour dosing interval have historically been flat; therefore, 4 samples over the interval were expected to allow reliable estimation of area under the concentration-time curve (AUC).
Plasma samples were assayed by validated liquid chromatography with tandem mass spectrometry (LC/MS/MS) methods. Duloxetine and the metabolites were quantified using 2 separate methods. The standard curve concentration range was 0.5 ng/mL to 100 ng/mL for duloxetine 31 and 1 ng/mL to 1000 ng/mL for the 2 metabolites.24,32
The pharmacokinetics of duloxetine and its 2 major metabolites were analyzed by standard noncompartmental methods using WinNonlin Professional Edition version 3.1 (Pharsight Corp., Cary, NC). The pharmacokinetic parameters for analysis were maximum steady-state plasma concentrations (Cmax) and the area under the concentration-time curve over a 12 hour dosing interval [AUC (0-12)]. Cmax, the sampling time of Cmax (tmax), and the minimum plasma concentration (Cmin) were observed from individual concentration-time data. AUC (0-12), average plasma concentration (Cavg), and apparent steady state plasma clearance (CL/F) were calculated using WinNonlin. The apparent terminal elimination rate constant was not calculated because the terminal phase was sampled for less than a single duloxetine half-life.
The sample size in this study was selected on the basis of prior clinical experience in ECG trials for other compounds. At least 90 subjects were planned to be enrolled so that at least 50 completed the trial. With 4 replicate QT measurements from 50 subjects, the study would have a 95% chance to conclude that the upper bound of 2-sided 90% confidence interval, for the mean difference between the duloxetine dose and placebo would be less than 10 msec, assuming that the true mean difference between duloxetine and placebo is 4.5 msec.
Statistical Methods to Evaluate QT Changes
For each subject at each time point, QT, RR, and QTc interval values were calculated as the average of 4 replicate values. The within-subject changes from baseline QT, RR, and QTc intervals for each treatment level at each scheduled time point were calculated by subtracting the corresponding baseline average values.
The treatment effects on QTc changes from baseline were analyzed by 3 QT correction methods for treatment effect. The first method was the mixed-effect analysis of covariance (ANCOVA) model with RR interval change from baseline as a covariate.33,34 The ANCOVA model also included treatment, time, and treatment-by-time interaction as the fixed effects; subject, subject-by-treatment, and subject-by-time interactions were included as the random effects. The least-squares mean change in the QTc interval was defined as the change in the QT interval corresponding to the zero change in the RR interval. The second correction method was the corrected QT interval using Fridericia's correction
The third correction method was the individual QT correction approach (QTcI35,36). The mixed-effect analysis of variance (ANOVA) model was used for these 2 correction approaches. The ANOVA model included treatment, time, and treatment-by-time interaction as the fixed effects; subject, subject-by-treatment, and subject-by-time interactions were included as the random effects. The individual correction QTcI for each subject was constructed in 2 steps: (1) fitting a linear model to the subject's drug-free replicate QT and RR values collected on the baseline day and in the placebo period, and (2) using the estimated slope to construct a linear subject-specific correction formula.
Subgroup analyses of QT data with subjects ≥40 years and <40 years of age were conducted using the above 3 correction methods. Outliers were defined as subjects who had an absolute magnitude of QTcF > 470 msec or a change from baseline in QTcF ≥30 msec, based on the average of 4 replicated individual QTcF values at each scheduled time point. The incidence of outliers for each treatment and level were summarized. Since there were only 5 patients aged over 65 years, no subgroup analysis was conducted.
All analyses were performed without adjustment for multiplicity. SAS 8.2 software (SAS Institute, Cary, NC) was used for all statistical analyses, and the mixed models were fit using the MIXED procedure. The restricted maximum likelihood estimation and the Kenward and Roger approximation of denominator degrees of freedom were specified in the mixed models.
Subject Demographics and Disposition
A total of 117 healthy female subjects aged 19 to 74 years participated in this study, with 56 subjects (49%) between the ages of 40 and 74 years (Table 1). Most subjects were white (59%). Out of 117 subjects, 116 had valid baseline evaluation and 70 subjects completed all protocol requirements. One subject had a positive drug screen and discontinued on day 33. Data from this subject were excluded from all further statistical analyses. During the study, 47 (40%) subjects discontinued for various reasons. Fifteen subjects discontinued because of adverse events, 18 discontinued due to physician's decision, 2 discontinued because of variance with the protocol, 1 discontinued because of the sponsor's decision, 1 died as a result of suicide judged to be unrelated to study drug treatment, and 10 discontinued due to subjects' decision. Descriptive and statistical analyses on QT interval included data from 116 subjects at baseline, 97 subjects on the fourth day of placebo dose level 3, 92 subjects on the fourth day of duloxetine 160 mg BID, 92 subjects on the fourth day of placebo dose level 4, 84 subjects on the fourth day of duloxetine 200 mg BID, and 97 subjects at moxifloxacin 400 mg.
Assay Sensitivity: Moxifloxacin Control
The mean changes in the QTc/QTcF/QTcI interval were significantly prolonged in subjects receiving moxifloxacin at both the 2 hour (6.3 msec, P < 0.001; 6.7 msec, P < 0.001; 6.6 msec, P < 0.001) and 6 hour (2.6 msec, P = 0.014; 2.7 msec, P = 0.019; 2.6 msec, P = 0.014) time points as compared with subjects in the placebo group (Table 2).
Comparison of Duloxetine to Placebo on QTc
Compared with placebo, the mean change in the QTc interval decreased with duloxetine 200 mg BID at 2, 6, 10, and 12 hours after dose by all 3 correction methods at each time point (Table 3). Furthermore, comparison of least-squares mean changes in the QTc interval between duloxetine 200 mg BID and its corresponding placebo dose level by these 3 correction methods demonstrated a decrease of at least 2.7 msec at each time point. Comparisons between duloxetine 160 mg BID and its corresponding placebo dose level showed similar results as those for 200 mg BID on the basis of the analyses of 3 different correction methods (Table 4). In all analyses of QTc, 3 correction methods resulted in the same conclusion of decrease in QT interval for 2 duloxetine doses compared with placebo. In addition, it is important to note that the range of the 90% confidence interval was consistently smallest for the ANCOVA approach. The tighter range of the 90% CI indicates that ANCOVA is a more precise method of estimating the mean change in the QT interval.
Subgroup Analysis by Age
The QT effects of duloxetine were analyzed separately for subjects who were at least 40 years old (Table 5). When compared with placebo, the mean change in QTc/QTcF/QTcI interval decreased at each time point with duloxetine 200 mg BID for subjects aged ≥40 years. The upper limit of the 90% CI was below 0 msec at all time points, indicating no clinically relevant increase in the QTc/QTcF/QTcI interval during duloxetine 200 mg BID compared with placebo.
No subject had a maximal QTc interval greater than 445 msec on the basis of the average of replicate QTcF values on the fourth day of duloxetine both at 160 mg BID and 200 mg BID. No subject experienced an increase in the QTcF interval from baseline greater than 60 msec on the basis of the average of replicate QTcF values. Four subjects from placebo dose level 3 and 2 subjects from placebo dose level 4 had a total of 8 increases in the QTcF interval from baseline >30 msec. The largest 2 QTcF increases from baseline were 50.8 msec and 42.5 msec for 10 and 12 hours, respectively, from the same subject after dosing at placebo dose level 3. The outlier values after duloxetine were 35.4, 30.7, and 31.9 msec from 2 subjects at 2, 6, and 10 hrs after dosing at 200 mg BID, respectively. Four subjects experienced QTcF increases from baseline >30 msec (36.4, 33.6, 31.3, and 32.1 msec, respectively) after taking one 400 mg dose of moxifloxacin.
Plasma concentration data were available from 93 subjects receiving duloxetine 160 mg BID and 84 subjects receiving 200 mg BID on the fourth day of each dose level. On the basis of the geometric mean estimates (Table 6), exposure to duloxetine was about 21% to 22% higher during dosing of 200 mg BID compared with 160 mg BID, which is consistent with the 25% increase in dose between those regimens. Duloxetine clearance was not different between these 2 doses. Consistent with previous studies at therapeutic doses, exposure to the 2 major metabolites exceeded by 2 times to the exposure to duloxetine.
Change in QTcF versus Concentration of Duloxetine or Metabolites
Scatter plots of the change in QTcF interval from baseline versus the plasma concentrations of duloxetine and the 2 major metabolites of duloxetine at the time of the ECG measurement are shown in Figures 2A, 2B. There was no discernable concentration effect for duloxetine (Figure 2A) or the 2 metabolites (Figures 2B and 2C) on the changes in QTcF interval from baseline. Similarly, neither duloxetine nor its metabolites had any effect on QTcI interval.
The adverse events reported by subjects receiving duloxetine were generally mild to moderate, but tolerability at high doses was limited by the additive effects of mild to moderate events over time. The most common adverse events were headache, nausea, insomnia, dizziness, somnolence, and constipation. A detailed safety assessment has been published in a companion paper elsewhere in this journal.37
In this clinically thorough QT/QTc study at supratherapeutic duloxetine doses, the mean QTc interval was not prolonged for any of the 3 correction methods. The results of this study are consistent with those obtained from previous clinical trials and hERG data suggesting that duloxetine, an SNRI, has less potential to induce an arrhythmia compared with other SNRIs such as venlafaxine and other antidepressants including selective serotonin reuptake inhibitors and tricyclic antidepressants.10,11,13,25,26,38
The in vitro studies showed that duloxetine moderately inhibits the hERG channel with an IC50 of 5.5 μM. Compounds that inhibit hERG are potentially torsadogenic for the heart. However, duloxetine in this study did not produce any QTc effects even at supratherapeutic doses. The underlying mechanism may not be apparent, but duloxetine may be allosterically modulating the hERG by interfering with other channel activities. For example, clozapine and verapamil are known blockers of hERG channel but lack torsadogenic potential for the heart.3 Both of these drugs are known calcium channel antagonists that may mitigate the hERG channel.3
Figure 3 illustrates that duloxetine concentrations in this study greatly exceeded those produced by the duloxetine dosage regimens typically used in clinical practice. At therapeutic doses, the median steady-state concentrations of duloxetine were 33.0 ng/mL and 69.1 ng/mL, respectively, during the dosing of 60 mg QD and 40 mg BID. Duloxetine Cmin and Cavg during dosing of 200 mg BID in the present study were more than 4 times and 5 times, respectively, the median concentration during dosing at 40 mg BID. These margins were even higher when compared with a dosing regimen of 60 mg QD, which produces a lower Cavg than does a regimen of 40 mg BID (Figure 3). The largest effect on duloxetine exposure is potent inhibition of CYP1A2, which increases duloxetine exposure by a factor of 5.39 The data show that the exposures produced by supratherapeutic doses in the present study are comparable to those that would be expected during coadministration of duloxetine and a potent CYP1A2 inhibitor.
An area of controversy regarding the clinical assessment of drug-induced QT interval prolongation concerns the methodology used to collect and evaluate digital electrocardiographic data. Generally, automated methods to evaluate digital ECGs have been criticized for their poor accuracy, and the preferred method has been manual over-reading of digital ECGs by a qualified cardiologist.40,41 To minimize interindividual variability in this study, all digital ECGs were over-read by a single cardiologist who was blinded to subject demographic data as well as treatment assignment.
Enrollment in this study was limited to women due to sex differences in duloxetine pharmacokinetics and QT effects in general. Women generally achieve higher duloxetine plasma concentrations than men, so enrolling only women ensured that the study population would have the highest possible exposure to duloxetine. Women also have been shown to have longer mean QT intervals and are usually considered to be at greater risk for torsades de pointes.42,43 On the basis of these 2 relationships, women would be expected to demonstrate more extensive QTc changes than men at a given supratherapeutic dose of duloxetine.
Although the association between increasing age and QTc prolongation is controversial, some studies have reported that both age and body mass index independently predict increases in the QTc interval in healthy subjects and suggested a possible mechanism for ventricular hypertrophy and myocardial action potential prolongation.44,45 The effect of age on QTc interval was investigated in the current analysis and found to have no apparent effect: QTc changes by any correction method for subjects aged ≥40 years were not different from the entire study population. As with younger subjects, when compared with placebo, the mean change in QTc interval decreased at all time points in subjects aged ≥40 years who received supratherapeutic doses of duloxetine.
These results demonstrated that duloxetine at supratherapeutic exposures in healthy females shortened the QTc interval as assessed by both mean changes and outliers in QT corrected by any correction method.
The authors acknowledge Alex Dmitrienko for his advice and careful review of this manuscript, Adelaine Yeo for statistical data analysis, and Lisa Ferguson-Sells for her help in preparing pharmacokinetic plots.
1. Fermini B, Fossa AA. The impact of drug-induced QT interval prolongation on drug discovery and development. Nat Rev Drug Discov
2. Roden DM. Drug-induced prolongation of the QT interval. N Engl J Med
3. Hoffmann P, Warner B. Are hERG channel inhibition and QT interval prolongation all there is in drug-induced torsadogenesis? A review of emerging trends. J Pharmacol Toxicol Methods
4. ICH Steering Committee. Safety pharmacology studies for assessing the potential for delayed repolarization (QT Interval Prolongation) by human pharmaceuticals. ICH Guidelines 57B Step 2; 2004.
5. ICH Steering Committee. The clinical evaluation of QT/QTc interval, prolongation and proarrythmic potential for non-antiarrythmic drugs. ICH Guidelines E14 Step 4; 2005.
6. Bymaster FP, Lee TC, Knadler MP, et al. The dual transporter inhibitor duloxetine: A review of its preclinical pharmacology, pharmacokinetic profile, and clinical results in depression. Curr Pharm Des
7. Kihara T, Ikeda M. Effects of duloxetine, a new serotonin and norepinephrine uptake inhibitor, on extracellular monoamine levels in rat frontal cortex. J Pharmacol Exp Ther
8. Wong DT, Bymaster FP, Mayle DA, et al. LY248686, a new inhibitor of serotonin and norepinephrine uptake. Neuropsychopharmacology
9. Berk M, Du PAD, Birkett M, et al. An open-label study of duloxetine hydrochloride, a mixed serotonin and noradrenaline reuptake inhibitor, in patients with DSM-III-R major depressive disorder. Int Clin Psychopharmacol
10. Detke MJ, Lu Y, Goldstein DJ, et al. Duloxetine, 60 mg once daily, for major depressive disorder: A randomized double-blind placebo-controlled trial. J Clin Psychiatry
11. Detke MJ, Lu Y, Goldstein DJ, et al. Duloxetine 60 mg once daily dosing versus placebo in the acute treatment of major depression. J Psychiatr Res
12. Detke MJ, Wiltse CG, Mallinckrodt CH, et al. Duloxetine in the acute and long-term treatment of major depressive disorder: a placebo- and paroxetine-controlled trial. Eur Neuropsychopharmacol
13. Goldstein DJ, Lu Y, Detke M, et al. Duloxetine in the treatment of depression: A double-blind placebo-controlled comparison with paroxetine. J Clin Psychopharmacol
14. Nemeroff CB, Schatzberg AF, Goldstein DJ, et al. Duloxetine for the Treatment of Major Depressive Disorder. Psychopharmacol Bull
15. Nelson JC, Wohlreich MM, Mallinckrodt CH, et al. Duloxetine for the treatment of major depressive disorder in older patients. Am J Ger Psychiatry
16. Goldstein DJ, Lu Y, Detke MJ, et al. Duloxetine vs. placebo in patients with painful diabetic neuropathy. Pain
17. Raskin J, Pritchett YL, Wang F, et al. A double-blind, randomized multicenter trial comparing duloxetine with placebo in the management of diabetic peripheral neuropathic pain. Pain Med
18. Raskin J, Pritchett YL, Chappell A, et al. Duloxetine in the treatment of diabetic neuropathic pain. Results from three clinical trials. Poster presented at the American Psychiatric Association on May 21, 2005 in Atlanta, GA.
19. Millard RJ, Moore K, Rencken R, et al. Duloxetine vs. placebo in the treatment of stress urinary incontinence: a four-continent randomized clinical trial. BJU International
20. Guay DRP. Duloxetine for management of stress urinary incontinence. Am J Geriatr Pharmacother
21. Westanmo AD, Gayken J, Haight R. Duloxetine: A balanced and selective norepinephrine- and serotonin-reuptake inhibitor. Am J Health Syst Pharm
22. Sharma A, Goldberg MJ, Cerimele BJ. Pharmacokinetics and safety of duloxetine, a dual-serotonin and norepinephrine reuptake inhibitor. J Clin Pharmacol
23. Lantz RJ, Gillespie TA, Rash TJ, et al. Metabolism, excretion, and pharmacokinetics of duloxetine in healthy human subjects. Drug Metab Dispos
24. Suri A, Reddy S, Gonzales C, et al. Duloxetine pharmacokinetics in cirrhotics compared with healthy subjects. Int J Clin Pharmacol Ther
25. Hudson JI, Wohlreich MM, Kajdasz DK, et al. Safety and tolerability of duloxetine in the treatment of major depressive disorder: Analysis of pooled data from eight placebo-controlled clinical trials. Hum Psychopharmacol
26. Thase ME. Effects of venlafaxine on blood pressure: a meta-analysis of original data from 3744 depressed patients. J Clin Psychiatry
27. Small D, Loghin C, Lucas R, et al. Pharmacokinetic evaluation of combine duloxetine and fluvoxamine dosing in CYP2D6 poor metabolizers. Clin Pharmacol Ther
28. Derby M, Gates M, Small D, et al. A definitive QT study incorporating step-wise dose escalation to steady-state. Clin Pharmacol Ther
29. Noel GJ, Natarajan J, Chien S, et al. Effects of three fluoroquinolones on QT interval in healthy adults after single doses. Clin Pharmacol Ther
30. Okin PM, Devereux RB, Howard BV, et al. Assessment of QT interval and QT dispersion for prediction of all-cause and cardiovascular mortality in American Indians: The Strong Heart Study. Circulation
31. Hua TC, Pan A, Chan C, et al. Effect of duloxetine on tolterodine pharmacokinetics in healthy volunteers. Br J Clin Pharmacol
32. McCulloch JD, Kuo F, Knadler MP, et al. Challenges during the development of a bioanalytical method for conjugated duloxetine metabolites in human plasma; chromatographic separation, sequestional SPE, and the use of positive and negative ion mass spectrometry. 51st ASMS Conference on Mass Spectromety and Allied Topics 2003.
33. Dmitrienko A, Smith BP. Analysis of QT interval in clinical trials. Drug Inf J
34. Dmitrienko A, Smith BP. Repeated-measures models in the analysis of QT interval. Pharm Stat
35. Batchvarov VN, Ghuran A, Smetana P, et al. QT-RR relationship in healthy subjects exhibits substantial intersubject variability and high intrasubject stability. Am J Physiol - Heart Circulatory Physiol
36. Malik M, Farbom P, Batchvarov V, et al. Relation between QT and RR intervals is highly individual among healthy subjects: implications for heart rate correction of the QT interval. Heart
37. Derby M, Zhang L, Chappell J, et al. Hemodynamic effects of duloxetine at supratherapeutic dosages. J Cardiovasc Pharmacol
. (Under Review).
38. Stahl SM, Grady MM, Moret C, et al. SNRIs: The pharmacology, clinical efficacy, and tolerability in comparison with other classes of antidepressants. CNS Spectrums
39. Preskorn SH. Duloxetine. J Psychiatr Pract
40. Johanson P, Armstrong PW, Barbagelata NA, et al. An academic ECG core lab perspective of the FDA initiative for digital ECG capture and data management in large-scale clinical trials. Drug Inf J
41. Strnadova C. The assessment of QT/QTc interval prolongation in clinical trials: a regulatory perspective. Drug Inf J
42. Makkar RR, Fromm BS, Steinman RT, et al. Female gender as a risk factor for torsades de pointes associated with cardiovascular drugs. J Am Med Assoc
43. Lindstrom E, Farde L, Eberhard J, et al. QTc interval prolongation and antipsychotic drug treatments: focus on sertindole. Int J Neuropsychopharmacol
44. Mangoni AA, Kinirons MT, Swift CG, et al. Impact of age on QT interval and QT dispersion in healthy subjects: a regression analysis. Age Ageing
45. Reardon M, Malik M. QT interval change with age in an overtly healthy older population. Clin Cardiol