QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.106.101881v1 318/1/352    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wisialowski, T. Articles by Fossa, A. A. Search for Related Content PubMed PubMed Citation Articles by Wisialowski, T. Articles by Fossa, A. A.

CARDIOVASCULAR

Differentiation of Arrhythmia Risk of the Antibacterials Moxifloxacin, Erythromycin, and Telithromycin Based on Analysis of Monophasic Action Potential Duration Alternans and Cardiac Instability

Pfizer Global Research and Development, Groton, Connecticut

Received January 26, 2006; accepted April 11, 2006.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
Antibacterial drugs are known to have varying degrees of cardiovascular liability associated with QT prolongation that can lead to the ventricular arrhythmia torsade de pointes. The purpose of these studies was to compare the assessment for the arrhythmogenic risk of moxifloxacin, erythromycin, and telithromycin. Each drug caused dose-dependent inhibition of the rapidly activating delayed rectifier potassium current encoded by the human ether-á-go-go-related gene (hERG) with IC₂₀ concentrations of 31 µM (moxifloxacin), 21 µM (erythromycin), and 11 µM (telithromycin). These drugs were also evaluated in an anesthetized guinea pig model to measure changes in monophasic action potential duration (MAPD) and to quantify beat-to-beat alternations in MAPD during rapid ventricular pacing. Moxifloxacin dose dependently increased MAPD and caused a rate-dependent increase in alternans at the highest achieved free drug concentration (41 µM). Erythromycin also increased MAPD at its highest free drug concentration (58 µM), but alternans occurred at a relatively lower therapeutic multiple (13.9 µM), and the magnitude of alternans at higher concentrations was independent of pacing rate. Further analysis of the data showed that the beat-to-beat pattern of alternans with erythromycin was less stable than that with moxifloxacin and suggestive of greater arrhythmogenic liability. In contrast to erythromycin and moxifloxacin, telithromycin decreased both MAPD and alternans at the highest achievable drug concentration (7.9 µM). The relative risk at therapeutic concentrations is erythromycin > moxifloxacin > telithromycin and appears to be consistent with clinical observations of torsade de pointes in patients.

Current preclinical regulatory guidelines for assessment of cardiovascular risk related to drug-induced QT interval prolongation include the use of both in vitro and in vivo assays (International Conference on Harmonisation, 2005a,b). As a consequence there are ongoing efforts to develop validated preclinical models that are useful to measure drug-induced prolongation and perhaps more importantly predict torsade de pointes (Friedrichs et al., 2005; Lawrence et al., 2005; Reinhart et al., 2005; Shah, 2005). There are increasing reports of QT prolongation and TdP associated with antibacterials (Frothingham, 2001; Iannini, 2002; Ray et al., 2004), and several models have been used to measure changes in cardiac repolarization and proarrhythmic liability with antibiotics such as: the Langendorff perfused rabbit heart (Milberg et al., 2002), methoxamine infused rabbit (Anderson et al., 2001; Akita et al., 2004), canine Purkinje fiber (Patmore et al., 2000), and the chronic atrioventricular-blocked dog (Chiba et al., 2004). We have recently reported results using an anesthetized guinea pig model to measure electrical alternans of monophasic action potential duration during rapid ventricular pacing with several hERG-blocking drugs (Fossa et al., 2004). Alternans is a measure of cardiac instability that is related to regulation of intracellular calcium handling and is thought to be a precursor or substrate to the development of ventricular arrhythmias (for review, see Clusin, 2003). We were able to quantify the magnitude of alternans and found differences between nonarrhythmogenic (i.e., verapamil) and arrhythmogenic (i.e., bepridil) hERG-blocking drugs even with similar pharmacology profiles at their respective clinical use levels.

In the present study we have evaluated three different antibacterials with known clinical outcomes in regards to arrhythmogenic risk: the fluoroquinolone moxifloxacin, the macrolide erythromycin, and the ketolide telithromycin. These drugs were examined for inhibition of the current encoded by hERG and subsequently tested in the alternans model to examine effects on instability of ventricular repolarization. These results were compared with published clinical outcomes at their relevant free drug concentrations.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Drugs
Erythromycin lactobionate was purchased from Abbott Laboratories (North Chicago, IL), moxifloxacin hydrochloride was purchased from Alchemie USA, Inc. (Plainville, CT), and telithromycin mesylate was synthesized by Pfizer, Inc. (Groton, CT). For voltage clamp experiments, all drugs were prepared as 5 to 100 mM dimethyl sulfoxide stock solutions and diluted as necessary. For the in vivo studies erythromycin was dissolved in sterile water, and both moxifloxacin and telithromycin were dissolved in 30% sulfobutylether cyclodextrin (SBE-CD) in distilled water. The cumulative doses for all drugs are shown in Table 1. Higher doses of telithromycin could not be achieved because of the limited solubility of this compound in SBE-CD; the solution for the highest dose of telithromycin was 107.52 mg/ml. Studies with vehicle solutions were conducted separately for comparison of drug effects.

Whole-Cell Voltage Clamp
The human embryonic kidney human embryonic kidney 293 cell line, recording solutions, and data analysis used in the hERG studies were described previously (Volberg et al., 2002). The standard external bath solution had the following ionic composition: 137 mM NaCl, 4 mM KCl, 1.8 mM CaCl₂, 1 mM MgCl₂, 10 mM glucose, and 10 mM HEPES, pH 7.4, with NaOH. The standard internal pipette solution was composed of: 130 mM KCl, 5 mM MgATP, 1.0 mM MgCl₂, 10mM HEPES, and 5 mM EGTA, pH 7.2, with KOH. All experiments were performed at 35 ± 1°C, and bath temperature was maintained by a temperature controller system (Warner Instruments, Hamden, CT).

Monophasic Action Potential Duration Alternans Measurements in the Anesthetized Guinea Pig
Surgical Procedure. All animal care conformed to the United States (Principles of Laboratory Animal Care, National Institutes of Health publication 85-23, revised 1996) and European guidelines for use of experimental animals. All studies were conducted in compliance with an Animal Care and Usage Protocol approved by the Institutional Animal Care and Use Committee at Pfizer, and these procedures are identical to those described previously (Fossa et al., 2004). In brief, guinea pigs were anesthetized with sodium pentobarbital, and a tracheotomy was performed for mechanical ventilation with room air. The right jugular vein and the left carotid artery were cannulated, and needle electrodes were placed for a lead II electrocardiogram. Pacing wires were sutured to the left ventricular apex, and monophasic action potential (MAP) signals were recorded from the left ventricular epicardium with an EPT 200 Probe (EP Technologies, San Jose, CA) that remained fixed in the same position throughout the entire study.

Dosing Protocol. Drugs or an equal volume of their respective vehicle (n = 6-8 each group) were administered i.v. via a programmable pump as an initial loading dose over 5 min followed by a maintenance infusion for 10 min. Blood samples were taken at the end of each dosing period, and the plasma was separated and frozen for future pharmacokinetic analysis.

Pacing Protocol. Pacing was conducted at twice the threshold current as follows: a 50-beat preconditioning pulse train (S1 = 220 ms or S1 = 240 ms) was followed immediately by a 30-beat pulse train at S2 = 200 ms. This S1-S2 protocol was repeated with a fixed S1 cycle length, whereas S2 decreased from 200 to 140 ms by 10-ms decrements unless an arrhythmia (TdP) occurred or the heart became refractory, in which case pacing was terminated. The protocol was repeated twice at baseline and during the last 5 min of each maintenance infusion period.

MAPD Measurements. MAPD₅₀ and MAPD₉₀ measurements were automatically performed by a Po-Ne-Mah system. Signals recorded during each study were visually inspected to verify ventricular capture and to validate measurements during the pacing period. Manual measurements of MAPD were made using a digital measure command on the Po-Ne-Mah when necessary.

Consecutive MAP signals that were accurately captured and measured during S2 pacing were used for alternans analysis. The first two paced beats at each cycle length were removed to account for possible incomplete capture. Beat-to-beat alternans was calculated as the absolute difference between consecutive MAPD₅₀ measurements and averaged over the entire S2 pulse train to yield average MAPD alternans. Data were included for a given study only if the difference in mean alternans between the two baseline periods indicated a stable experimental preparation: 5 ms for S2 = 200 - 170; 10 ms for S2 = 160; and 15 ms for S2 = 150 - 140.

Further inspection of the data revealed phase reversal patterns of beat-to-beat alternans (e.g., short-long-short pattern changed to long-short-short-long; for example, see Fig. 5, A-C) that we believe are suggestive of greater cardiac instability and increased arrhythmogenic liability. These unstable patterns were observed with the higher doses of moxifloxacin and erythromycin but did not occur with telithromycin at any dose level. To quantify this phenomenon, we classified a phase reversal as a change in this short-long-short pattern during alternans greater than 10 ms.

View larger version (32K): [in this window] [in a new window]   Fig. 5. Example of the beat-to-beat MAP signal, MAPD50, and alternans during infusion of erythromycin (58.3 µM). A, beat-to-beat raw MAP signals from the end of pacing at S1 = 220 ms, followed by the pulse train at S2 = 160 ms, and then the return to sinus rhythm. B, beat-to-beat MAPD50 values during pacing at S2 = 160 ms. C, beat-to-beat alternans during this same period. Stable MAP signals during pacing at S1 = 220 ms and during sinus rhythm are denoted by (N). During pacing at S2 = 160 ms MAPD50 alternates between long (L) and short (S) duration except where the phase reversals are present in the middle of the pulse train. The phase reversals are identified in A and B by * and in C by solid arrows.

Plasma Protein Binding. Plasma protein binding was determined by equilibrium dialysis (96-well microequalibrium dialysis device; HTDialysis, LLC, Gales Ferry, CT). Dialysis membranes had a molecular mass cutoff of 12,000 to 14,000 Da. Plasma protein binding for erythromycin and telithromycin were examined at 0.4, 4, 16, and 64 µg/ml, and moxifloxacin plasma protein binding was examined at 1, 4, 16, and 64 µg/ml. In brief, plasma and buffer samples (100 µl) were quenched with internal standard (300 µl of midazolam and 500 ng/ml in acetonitrile) after equilibration. The samples were then centrifuged, and the supernatant was analyzed by high-performance liquid chromatography/MS/MS as described above.

Expression of Results and Statistical Analyses. Blood pressure, heart rate, and MAPD data reported in the body of this article and in Table 2 are expressed as differences from baseline. The alternans (see data shown Figs. 2, 3, 4) are expressed as differences from time-matched vehicle responses. For statistical analyses the effects of moxifloxacin and telithromycin were compared with the vehicle SBE-CD and erythromycin was compared with the vehicle sterile water. Statistical analysis of average MAPD, alternans, heart rate, and blood pressure was performed using SAS version 8 (SAS Institute, Cary, NC), and the detailed methods have been described previously (Fossa et al., 2004). Statistical analysis to compare the relative frequency of phase reversals observed with moxifloxacin and erythromycin was done using the Wilcoxon signed rank test in Stat-View, version 5.0.1. Comparisons of moxifloxacin or erythromycin to its respective vehicle using this test could not be done because there was no signal with vehicle treatment (i.e., phase reversal count = 0). Statistical significance for all analyses was set at P < 0.05.

View larger version (23K): [in this window] [in a new window]   Fig. 2. Mean MAPD50 alternans (ALT) with moxifloxacin (n = 6). Anesthetized guinea pigs were treated with an i.v. infusion of moxifloxacin to achieve free drug plasma concentrations of 0.8, 3.1, 10.0, and 41.0 µM. The clinical multiple for each drug at each dose is also listed in parentheses. Statistical significance difference from predose baseline compared with the respective vehicle response at each dose level (P < 0.05) is noted by *.

View larger version (24K): [in this window] [in a new window]   Fig. 3. Mean MAPD50 alternans (ALT) with erythromycin (n = 8). Anesthetized guinea pigs were treated with intravenous infusion of erythromycin to achieve free drug plasma concentrations of 0.8, 3.0, 13.9, and 58.3 µM. The clinical multiple for each drug at each dose is also listed in parentheses. Statistical significance difference from predose baseline compared with the respective vehicle response at each dose level (P < 0.05) is noted by *.

View larger version (21K): [in this window] [in a new window]   Fig. 4. Mean MAPD50 alternans (ALT) with telithromycin (n = 7). Anesthetized guinea pigs were treated with an i.v. infusion of telithromycin to achieve free drug plasma concentrations of 0.1, 0.4, 2.3, and 7.9 µM. The clinical multiple for each drug at each dose is also listed in parentheses. Statistical significance difference from predose baseline compared with respective vehicle response at each dose level (P < 0.05) is noted by *.

	Results

Top Abstract Materials and Methods Results Discussion References

 
In Vitro hERG Inhibition. The effects of moxifloxacin, erythromycin, and telithromycin on hERG current recorded in HEK293 cells are shown in Fig. 1, and a representative example of recordings obtained in the absence and presence of moxifloxacin is presented in Fig. 1A. For each antibiotic studied, hERG current inhibition was measured (n = 4-5 cells) at concentrations ranging from 10 to 300 µM. Drugs were perfused in the experimental chamber in a cumulative manner, and dose-response relationships were determined by fitting the data with a Hill equation to obtain inhibitory concentration values of 20 and 50% (IC₂₀ and IC₅₀). Although hERG IC₅₀ values are traditionally used for comparison of data, previous studies by our group (Fossa et al., 2004) and others (Redfern et al., 2003; Jonker et al., 2005) have shown that clinically significant prolongation of the QT interval is generally associated with inhibition of hERG current amplitude ranging from 10 to 20%. The IC₂₀ and IC₅₀ values determined for the three antibiotics studied were moxifloxacin, 31 and 102 µM, erythromycin, 21 and 96 µM, and telithromycin, 11 and 46 µM, respectively.

View larger version (17K): [in this window] [in a new window]   Fig. 1. Effects of moxifloxacin (MOX), erythromycin (ERY), and telithromycin (TEL) on hERG current amplitude. A, superimposed current traces recorded in the absence (0) and presence of 30, 100, and 300 µM MOX (1, 2, and 3, respectively). Currents were elicited by a 1-s depolarization to +20 mV from a holding potential of -80 mV, followed by a repolarizing ramp (0.5 V/s) to -80 mV to mimic repolarization of an action potential. The stimulation frequency was 0.25 Hz. B, concentration-response relationships obtained for MOX, ERY, and TEL (n = 4-5 cells each drug). The steady-state inhibition is plotted against drug concentration. Data were fitted using a Hill equation.

Pharmacokinetic Analysis. The infusion protocols for moxifloxacin, erythromycin, and telithromycin resulted in mean free drug concentrations of 41.0 ± 8.1, 58.3 ± 4.9, and 7.9 ± 0.6 µM, respectively, at the highest dose level. For comparison, we calculated the ratio of these free drug concentrations to their respective clinical efficacious free drug concentrations (C_eff) as reported in the literature, and in the case of erythromycin we used the value that is obtained during intravenous administration as this regimen yields higher drug concentrations and a greater incidence of torsade de pointes. The highest free drug concentrations achieved in our studies exceeded C_eff by approximately 8-fold and are summarized in Table 1. The free drug/C_eff and free drug/hERG IC₂₀ ratios achieved are also listed for comparison. The highest concentrations achieved for moxifloxacin, erythromycin, and telithromycin yielded free drug/hERG IC₂₀ ratios of 1.32, 2.78, and 0.72, respectively. The solubility of telithromycin limited infusion of higher doses of drug. In addition, given the relatively higher molecular weight and lower free fraction in plasma of telithromycin, this unfortunately resulted in a lower free drug/hERG IC₂₀ ratio compared with those for moxifloxacin and erythromycin.

Heart Rate, Blood Pressure, and MAPD at Rest. There were no statistical differences between the SBE-CD and sterile water vehicle solutions as summarized in Table 2. The effects of moxifloxacin, erythromycin, and telithromycin on blood pressure, heart rate, and MAPD are also summarized in Table 2. Moxifloxacin and erythromycin decreased heart rate at the highest concentrations by 67 and 83 bpm, respectively, compared with decreases of 23 and 25 bpm with their respective vehicles. Telithromycin at 2.3 µM produced a smaller decrease in heart rate (7 bpm) than its SBE-CD vehicle (22 bpm). All compounds decreased blood pressure at their highest drug concentrations (moxifloxacin, 13 mm Hg; erythromycin, 26 mm Hg; and telithromycin, 16 mm Hg) compared with vehicle responses ranging between a decrease of 2 mm Hg and an increase of 4 mm Hg. Erythromycin also significantly decreased blood pressure at 13.9 µM by 12 mm Hg. Moxifloxacin at 41 µM increased resting MAPD₉₀ and MAPD₅₀ by 59 and 54 ms, compared with respective SBE-CD vehicle responses of 20 and 23 ms. Erythromycin increased resting MAPD₉₀ and MAPD₅₀ by 71 and 48 ms at 58 µM compared with respective water vehicle responses of 22 and 27 ms. Telithromycin caused relatively smaller increases in MAPD₉₀ (8 ms at 2.3 µM and 14 ms at 7.9 µM) and MAPD₅₀ (11 ms at 7.9 µM) compared with the respective vehicle responses.

MAPD during Ventricular Pacing. The effects of these drugs on MAPD during ventricular pacing at S2 = 200 ms are also shown in Table 2. Moxifloxacin increased MAPD₉₀ and MAPD₅₀ by 23 and 22 ms, respectively, at 41 µM compared with its SBE-CD vehicle responses of 11 and 13 ms. Erythromycin significantly increased MAPD₉₀ at 13.9 and 58 µM by 14 and 22 ms, respectively, compared with time-matched vehicle responses of 8 and 11 ms. Erythromycin caused a significantly smaller increase in MAPD₅₀ at 58 µM (8 ms prolongation) compared with a 17-ms increase with its water vehicle response. Likewise, telithromycin at 7.9 µM caused significantly smaller increases in MAPD₉₀ and MAPD₅₀ of 3 and 2 ms compared with respective vehicle increases of 11 and 13 ms.

MAPD Alternans during Ventricular Pacing. The effects of moxifloxacin, erythromycin, and telithromycin on MAPD₅₀ alternans are illustrated in Figs. 2, 3, 4; the effect of each drug is plotted relative to the appropriate vehicle response. At the highest concentration, moxifloxacin caused a large rate-dependent increase in alternans of 5 to 38 ms at cycle lengths 180 ms and below (Fig. 2). In comparison, erythromycin at 58 µM increased alternans by 15 to 20 ms over the full range of cycle lengths from 200 to 140 ms. A more rate-dependent effect was observed with erythromycin at 13.9 µM beginning at S2 = 160 ms (Fig. 3). Finally, telithromycin at 7.9 µM produced a small decrease in alternans at the fastest cycle length S2 = 140 ms, indicative of a cardiac stabilizing activity (Fig. 4).

Phase Reversal during MAPD Alternans. An example of the raw MAP signal recorded during infusion of erythromycin (58.3 µM) is shown in Fig. 5A. These beat-to-beat signals are from the end of the S1 = 220 ms pulse train, followed by pacing at S2 = 160 ms, and then upon return to sinus rhythm. The pattern during S1 = 220 ms is stable (denoted N), but once the pacing rate is accelerated to S2 = 160 ms, MAPD alternans develops between long (L) and short (S) on successive beats. In the middle of the pulse train there are two instances for which the pattern of L-S-L-S is interrupted, and there is a phase reversal (denoted by *). Finally, the alternans pattern of L-S-L-S resumes toward the end of pacing period. The MAP signal becomes stable (N) immediately upon return to sinus rhythm. The corresponding values for MAPD₅₀ and alternans during the period of measurement (S2 = 160 ms) are shown in Fig. 5, B and C, respectively. Note the varying pattern of MAPD₅₀ and the magnitude of alternans from the beginning to the end of this record. Although the largest amount of alternans is nearly 80 ms, alternans is reduced during the period of phase reversal and the result is a decrease in mean alternans. The analysis of phase reversals in the beat-to-beat alternans data was performed during the higher doses of moxifloxacin and erythromycin by counting the frequency of these occurrences within each study. The results are summarized in Fig. 6, which illustrates the proportion of animals that displayed one or more phase reversals at each S2 cycle length during infusion of either erythromycin or moxifloxacin. Statistical analysis of these data showed that erythromycin caused an increase in the percentage of animals with a phase reversal during alternans compared with moxifloxacin at both dose 3 and dose 4 (P < 0.05).

View larger version (23K): [in this window] [in a new window]   Fig. 6. Summary results from the analysis of alternans phase reversal: percentage of animals treated with moxifloxacin (MOX) (n = 6) or erythromycin (ERY) (n = 8) having the appearance of one or more phase reversals that occurred during pacing at each S2 cycle length after dose 3 and dose 4 of MOX (10.0 and 41.0 µM) or ERY (13.9 and 58.3 µM). *, P < 0.05, ERY versus MOX dose 3; , P < 0.05, ERY versus MOX dose 4.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
These studies demonstrate the utility of the anesthetized guinea pig alternans model to differentiate the potential proarrhythmic effects of these antibacterials with diverse hERG inhibition profiles relative to their clinically used free drug concentrations. At the maximal achieved free drug concentrations that were 7- to 8-fold clinical efficacious concentrations, moxifloxacin and erythromycin increased ventricular repolarization and increased MAPD alternans. Moxifloxacin caused a greater magnitude of alternans with similar increases in MAPD at 8.2-fold C_eff (1.3-fold hERG IC₂₀). Erythromycin increased alternans at 1.6-fold C_eff, yielding a lower therapeutic index. In addition, it increased alternans over a wider range of cycle lengths and caused a relatively less stable pattern of alternans at 6.9-fold C_eff (2.8-fold hERG IC₂₀). Telithromycin did not increase either MAPD or alternans up to 7.9-fold C_eff (0.7-fold hERG IC₂₀). This comparative relationship (erythromycin > moxifloxacin > telithromycin) is similar to the reported cases of torsade de pointes for these drugs, although telithromycin has only been in the marketplace since 2001 and may not have a definitive cardiovascular safety record.

Despite causing modest QT prolongation in humans, moxifloxacin has caused relatively few clinical cases of torsade de pointes. Data from the Spontaneous Reporting System (1969-1997) and the Adverse Event Reporting System show that 0 cases of torsade de pointes per 10 million prescriptions were reported (Frothingham, 2001) and in a postmarketing safety evaluation study required by the FDA there was 1 case per 1.2 million patients who also had other risk factors associated with torsade de pointes (Owens, 2004). Moxifloxacin increased sinus rhythm and paced MAPD₉₀ and MAPD₅₀ at a concentration 8-fold above C_eff and also increased alternans in a rate-dependent fashion from cycle lengths of 180 to 140 ms. Moxifloxacin has also been reported to increase MAPD₉₀ and MAPD₅₀ at concentrations of 10 µM (2-fold C_eff) and higher in an isolated-perfused rabbit wedge preparation, and moxifloxacin at concentrations between 30 and 100 µM (6-20-fold C_eff) increased the incidence of early afterdepolarizations (Chen et al., 2005). Other studies have shown that moxifloxacin caused torsade de pointes and death in atrioventricular blocked dogs at 15 µM free drug (12 µg/ml with 50% protein binding) (Chiba et al., 2004). Over the range of its clinical use concentration (1-5 µM free drug), moxifloxacin is reported to increase QTc by 5 ms for every 1000 mg/ml (1.25 µM free drug; FDA Advisory Committee Hearing), even though this is considerably below the hERG IC₂₀ of 31 µM. However, the safety record of moxifloxacin is quite favorable and is related to several important factors. There are no known significant cytochrome P450 interactions with moxifloxacin. Because moxifloxacin is excreted unchanged, there are also no complications in renal or hepatic impaired patients (Avelox package insert, Bayer HealthCare, West Haven, CT). Thus, the pharmacokinetics of moxifloxacin are very well controlled, and this is probably a major reason few cases of torsade de pointes have been observed or reported.

Erythromycin, on the other hand, causes greater QT prolongation in humans and has a higher incidence rate of torsade de pointes. The highest plasma concentration achieved in our studies, 58.3 µM, represents a 6.9-fold increase above intravenous C_eff and 2.8-fold the hERG IC₂₀. At this concentration MAPD₉₀ was significantly increased during both sinus rhythm and ventricular pacing. Interestingly, the increase in MAPD₉₀ was larger compared with MAPD₅₀ during sinus rhythm, and MAPD₅₀ was decreased during ventricular pacing. These nonuniform changes in action potential duration are indicative of triangulation, a property that has also been described as a potential risk for drug-induced arrhythmia (Hondeghem et al., 2001). In our studies alternans was also significantly increased at concentrations equal to 1.6- and 6.9-fold C_eff.

As previously described, erythromycin increased the occurrence of phase reversals in the pattern of MAPD alternans compared with moxifloxacin. A similar analysis with T-wave alternans has been reported as a univariate predictor of survival in patients with ischemic left ventricular dysfunction (Narayan et al., 2005). Our group has previously reported phase reversal of MAPD₅₀ as discordant alternans with the class III antiarrhythmic E-4031 (1-[2-(6-methyl-2-pyridyl)ethyl]-4-methyl-sulfonylaminobenzoyl)-piperidine) (Fossa et al., 2004). The occurrence of this phenomenon is suggestive of proarrhythmic activity, in which the beat-to-beat sequence of MAPD alternans in one isolated region of the heart is out of phase with one or more neighboring regions (Pastore et al., 1999). An increase in repolarization heterogeneity between epicardial electrograms, including discordant alternans, has also been shown to occur prior to ventricular fibrillation during coronary occlusion in anesthetized dogs (Nearing and Verrier, 2003). The analysis of beat-to-beat measures of APD (Hondeghem et al., 2001) and QT interval duration (Schneider et al., 2005; Van der Linde et al., 2005) has also been used as additional measures of drug-induced repolarization instability and arrhythmogenic liability.

In humans, erythromycin has been reported to increase the QT interval by 8 to 31 ms (Iannini, 2002) and cause fatal arrhythmias when given alone (incidence-rate ratio of 2.01 or 10 deaths/5305 person-years) and in combination with other QT-prolonging drugs (Ray et al., 2004). Erythromycin undergoes extensive metabolism and the incidence rate ratio of sudden cardiac death appears to be concentration-related. Plasma concentrations increase 2.5-fold when given with inhibitors of cytochrome P450_3A4 and 10-fold when given alone intravenously (De Bruin et al., 2005). Erythromycin alters the metabolism of other compounds via the cytochrome P450 system, resulting in elevated plasma concentrations of other drugs as well that could also lead to excessive QT prolongation and cardiac arrhythmias (FDA Briefing Package). Our results with erythromycin are in agreement with an increased risk of QT prolongation and torsade de pointes.

The ketolide telithromycin has a relatively good cardiac safety record, although it has only been in the marketplace since 2001. The hERG IC₂₀ for telithromycin (11 µM) yields a hERG therapeutic index of 11-fold that is comparatively higher than those for moxifloxacin (6.2-fold) and erythromycin (2.5-fold). At concentrations as high as 7.9 µM, telithromycin did not increase MAPD₉₀ or alternans but caused slight decreases in both measures. However, telithromycin has been reported to increase APD duration in isolated canine Purkinje fibers at concentrations near 20% hERG inhibition and above (Martin et al., 2004). Data in the FDA Briefing Package also indicate that telithromycin at 1 µM increased APD₅₀ and APD₉₀ in isolated rabbit Purkinje fibers by 11 and 7%, respectively. Phase I clinical trials showed that telithromycin increased QTc by as much as 28 ms at a dose of 2400 mg, whereas other data suggest that telithromycin has no effect on the QT interval at therapeutic concentrations (800 mg) (Demolis et al., 2003). In addition to these effects of telithromycin by itself, drug-drug interaction studies with telithromycin increased plasma concentrations of cisapride, and this combination increased QTc by approximately 30 ms compared with 10 ms with either drug alone. Concentrations of telithromycin itself also vary with cytochrome 3A4 inhibition, in the elderly and in patients with impaired renal function. Therefore, it is possible that telithromycin concentrations observed in humans could exceed those tested in this study leading to further QT prolongation and arrhythmia. As previously mentioned, there were several factors that limited the telithromycin concentrations in our studies to 7.9-fold C_eff and a free drug/hERG IC₂₀ ratio of 0.72. A comparison could be made to the third dose of erythromycin that achieved a free drug/hERG IC₂₀ ratio of 0.66. At this concentration erythromycin increased mean alternans by nearly 20 ms at S2 = 150 ms (refer to Fig. 3) compared with telithromycin, which had negligible effects (Fig. 4). Nonetheless, at roughly 8-fold clinical efficacious concentrations telithromycin appears to have less arrhythmogenic risk than moxifloxacin and erythromycin.

Limitations. Although the known cardiovascular risks associated with these and other drugs that cause torsade de pointes are mostly associated with inhibition of hERG, it is important to recognize that alternans is a function of intracellular calcium handling. However, because of the toxicity of the agents required to image intracellular calcium, it is not possible to conduct these measurements in situ. It would be of considerable interest to conduct similar studies using isolated cardiac tissues or myocytes to determine whether calcium handling is altered with these compounds.

Our results are based on extracellular recordings made from a single location on the left ventricular epicardium. Likewise, the analysis and interpretation of phase reversals and the association with discordant alternans are based on literature suggesting that this pattern is consistent with increased heterogeneity of repolarization throughout the heart. To more accurately measure this phenomenon it would be necessary to have multiple recording sites to definitively show this instability because spatial heterogeneity has been demonstrated with erythromycin in canine left ventricular tissue (Antzelevitch et al., 1996).

In summary, these measures of cardiac instability observed during ventricular pacing are predictive of the relative arrhythmogenic risk of these antibacterials: erythromycin > moxifloxacin > telithromycin. In addition to hERG inhibition and QT or MAPD prolongation, quantification of these parameters could lead to improved cardiovascular risk assessment for current and new drugs.

	Acknowledgements

 
It is with much gratitude that we acknowledge the contributions of William Gorczyca, Roxanne L. Winslow, and Eric Wolfgang for conducting the experiments using the anesthetized guinea pig and Jianlin Feng, Walter Volberg, and Shuya Wang for performing the voltage-clamp experiments.

	Footnotes

 
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.

doi:10.1124/jpet.106.101881.

ABBREVIATIONS: hERG, human ether-á-go-go-related gene, TdP, torsade de pointes like arrhythmia; SBE-CD, sulfobutylether cyclodextrin; MAP, monophasic action potential; MAPD, MAP duration; FDA, Food and Drug Administration; QTc, rate-corrected QT interval; E-4031, 1-[2-(6-methyl-2-pyridyl)ethyl]-4-methyl-sulfonylaminobenzoyl)-piperidine; APD, action potential duration.

Address correspondence to: Dr. Anthony A. Fossa, Pfizer Global Research and Development, Eastern Point Rd., Building 274, Groton, CT 06340. E-mail: anthony.a.fossa{at}pfizer.com

	References

Top Abstract Materials and Methods Results Discussion References

 

Akita M, Shibazaki Y, Izumi M, Hiratsuka K, Sakai T, Kurosawa T, and Shindo Y (2004) Comparative assessment of prurifloxacin, sparfloxacin, gatifloxacin and levofloxacin in the rabbit model of proarrhythmia. J Toxicol Sci 29: 63-71.[Medline]

Anderson ME, Al-Khatib SM, Roden DM, and Califf RM (2002) Cardiac repolarization: current knowledge, critical gaps and new approaches to drug development and patient management. Am Heart J 144: 769-781.[CrossRef][Medline]

Anderson ME, Mazur A, Yang T, and Roden DM (2001) Potassium current antagonist properties of proarrhythmic consequences of quinolone antibiotics. J Pharmacol Exp Ther 296: 806-810.[Abstract/Free Full Text]

Antzelevitch C, Sun Z, Zhang Z, and Yan G (1996) Cellular and ionic mechanisms underlying erythromycin-induced long QT intervals and torsade de pointes. J Am Coll Cardiol 28: 1836-1848.[Abstract]

Chen X, Cass J, Bradley J, Dahm C, Sun Z, Kadyszewski E, Engwall M, and Zhou J (2005) QT prolongation and proarrhythmia by moxifloxacin: concordance of preclinical models in relation to clinical outcome. Br J Pharmacol 146: 1-8.[CrossRef][Medline]

Chiba K, Sugiyama A, Hagiwara T, Takahashi S, Takasuna K, and Hashimoto K (2004) In vivo experimental approach for the risk assessment of fluoroquinolone antibacterial agents-induced long QT syndrome. Eur J Pharmacol 486: 189-200.[Medline]

Clusin WT (2003) Calcium and cardiac arrhythmias: DADs, EADs and alternans. Crit Rev Clin Lab Sci 40: 337-375.[Medline]

De Bruin ML, Pettersson M, Meyboom RHB, Hoes AW, and Leufkens HGM (2005) Anti-hERG activity and the risk of drug-induced arrhythmias and sudden death. Eur Heart J 26: 590-597.[Abstract/Free Full Text]

Demolis JL, Vacheron F, Cardus S, and Funck-Brentano C (2003) Effect of a single and repeated oral doses of telithromycin on cardiac QT interval in healthy subjects. Clin Pharmacol Ther 73: 242-252.[CrossRef][Medline]

Fenichel RR, Malik M, Antzelevitch C, Sanguinetti M, Roden DM, Priori SG, Ruskin JN, Lipicky RJ, and Cantilena LR (2004) Drug-induced torsades de pointes and implications for drug development. J Cardiovasc Electrophysiol 15: 475-495.[CrossRef][Medline]

Fossa AA, Wisialowski T, Wolfgang E, Wang E, Avery A, Raunig RL, and Fermini B (2004) Differential effect of hERG blocking agents on cardiac electrical alternans in guinea pig. Eur J Pharmacol 486: 209-221.[CrossRef][Medline]

Friedrichs GS, Patmore L, and Bass A (2005) Non-clinical evaluation of ventricular repolarization (ICH S7B): results of an interim survey of international pharmaceutical companies. J Pharmacol Toxicol Methods 52: 6-11.[Medline]

Frothingham R (2001) Rates of torsades de pointes associated with ciprofloxacin, ofloxacin, levofloxacin, gatifloxacin and moxifloxacin. Pharmacotherapy 21: 1468-1472.[CrossRef][Medline]

Hondeghem LM, Carlsson L, and Duker G (2001) Instability and triangulation of the action potential predict serious proarrhythmia, but action potential duration prolongation is antiarrhythmic. Circulation 103: 2004-2013.[Abstract/Free Full Text]

Iannini PB (2002) Cardiotoxicity of macrolides, ketolides and fluoroquines that prolong the QTc interval. Expert Opin Drug Saf 1: 121-128.[Medline]

International Conference on Harmonisation of technical requirements for registration of pharmaceuticals for human use. ICH Harmonised Tripartite Guidelines (2005a) The clinical evaluation of QT/QTc interval prolongation and proarrhythmia potential for non-antiarrhythmic drugs. E14. Accessed 12 May 2005. http://www.ich.org/LOB/media/MEDIA1476.pdf.

International Conference on Harmonisation of technical requirements for registration of pharmaceuticals for human use. ICH Harmonised Tripatite Guidelines (2005b) The non-clinical evaluation of the potential for delay ventricular repolarization (QT interval prolongation) by human pharmaceuticals. S7B. Accessed 12 May 2005. http://www.ich.org/LOB/media/MEDIA2192.pdf.

Jonker DM, Kenna LA, Leishman D, Wallis R, Milligan PA, and Jonsson EN (2005) A pharmacokinetic-pharmacodynamic model for the quantitative prediction of dofetilide clinical QT prolongation from human ether-á-go-go-related gene current inhibition data. Clin Pharmacol Ther 75: 572-582.

Kang J, Wang L, Chen XL, Triggle DJ, and Rampe D (2001) Interactions of a series of fluoroquinolone antibacterial drugs with the human cardiac K⁺ channel hERG. Mol Pharmacol 59: 122-126.[Abstract/Free Full Text]

Lawrence CL, Pollard CE, Hammond TG, and Valentin JP (2005) Nonclinical proarrhythmia models: predicting torsades de pointes. J Pharmacol Toxicol Methods 52: 46-59.[CrossRef][Medline]

Martin RL, McDermott JS, Salmen HJ, Palmatier J, Cox BF, and Gintant GA (2004) The utility of hERG and repolarization assays in evaluating delayed cardiac repolarization: influence of multi-channel block. J Cardiovasc Pharmacol 43: 369-379.[CrossRef][Medline]

Milberg P, Eckardt L, Bruns HJ, Biertz J, Ramtim S, Reinsch N, Fleischer D, Kirchhof P, Fabritz L, Breithardt G, et al. (2002) Divergent proarrhythmic potential of macrolide antibiotics despite similar QT prolongation: fast phase 3 repolarization prevents early afterdepolarizations and torsade de pointes. J Pharmacol Exp Ther 303: 218-225.[Abstract/Free Full Text]

Narayan SM, Smith JM, Schechtman KB, Lindsay BD, and Cain ME (2005) T-wave alternans following ventricular extrasystoles predicts arrhythmia-free survival. Heart Rhythm 2: 234-241.[CrossRef][Medline]

Nearing BD and Verrier RL (2003) Tracking cardiac electrical instability by computing interlead heterogeneity of T-wave morphology. J Appl Physiol 95: 2265-2272.[Abstract/Free Full Text]

Owens RC (2004) QT prolongation with antimicrobial agents: understanding the significance. Drugs 64: 1091-1124.[CrossRef][Medline]

Pastore JM, Girourd SD, Laurita KR, Akar FG, and Rosenbaum DS (1999) Mechanism linking T-wave alternans to the genesis of cardiac fibrillation. Circulation 99: 1385-1394.[Abstract/Free Full Text]

Patmore L, Fraser S, Mair D, and Templeton A (2000) Effects of sparfloxacin, grepafloxacin, moxifloxacin and ciprofloxacin on cardiac action potential duration. Eur J Pharmacol 406: 449-452.[Medline]

Ray WA, Murray KT, Meredith S, Narasimhulu SS, Hall K, and Stein CM (2004) Oral erythromycin and the risk of sudden death from cardiac causes. N Engl J Med 351: 1089-1096.[Abstract/Free Full Text]

Redfern WS, Carlsson L, Davis AS, Lynch WG, MacKenzie I, Palethorpe S, Siegl PKS, Strang I, Sullivan AT, Wallis R, et al. (2003) 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 58: 32-45.[Abstract/Free Full Text]

Reinhart GA, Fryer RM, Osinski MA, Polakowski JS, Cox BF, and Gintant GA (2005) Predictive, non-GLP models of secondary pharmacodynamics: putting the best compounds forward. Curr Opin Chem Biol 9: 392-399.[Medline]

Schneider J, Hauser R, Andreas JO, Linz K, and Jahnel U (2005) Differential effects on human ether-á-go-go-related gene (hERG) blocking agents on QT duration variability in conscious dogs. Eur J Pharmacol 512: 53-60.[CrossRef][Medline]

Shah RR (2005) Drugs, QT interval prolongation and ICH E14: the need to get it right. Drug Safety 28: 115-125.[CrossRef][Medline]

Van der Linde H, Van de Water A, Loots W, Van Deuren B, Lu HR, Van Ammel K, Peeters M, and Gallacher DJ (2005) A new method to calculate the beat-to-beat instability of QT duration in drug-induced long QT in anesthetized dogs. J Pharmacol Toxicol Methods 52: 168-177.[CrossRef][Medline]

Volberg WA, Koci BJ, Su W, Lin J, and Zhou J (2002) Blockade of human cardiac potassium channel human ether-á-go-go related gene (hERG) by macrolide antibiotics. J Pharmacol Exp Ther 302: 320-327.[Abstract/Free Full Text]

This article has been cited by other articles:

				A. A. Fossa, T. Wisialowski, J. N. Duncan, S. Deng, and M. Dunne Azithromycin/Chloroquine Combination Does Not Increase Cardiac Instability despite an Increase in Monophasic Action Potential Duration in the Anesthetized Guinea Pig Am J Trop Med Hyg, November 1, 2007; 77(5): 929 - 938. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.106.101881v1 318/1/352    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wisialowski, T. Articles by Fossa, A. A. Search for Related Content PubMed PubMed Citation Articles by Wisialowski, T. Articles by Fossa, A. A.