Drug Interaction Between Mosapride and Erythromycin Without Electrocardiographic Changes

Takao KATOH,1 MD, Hirokazu SAITOH,1 MD, Norihiko OHNO,1 MD, Masao TATENO,2 MD, Tsuyoshi NAKAMURA,2 MD, Isao DENDO, 2 MD, Shinichi KOBAYASHI,3 MD, and Koichi NAGASAWA,4 MD

SUMMARY QT prolongation and torsades de pointes have been documented in patients administered cisapride and its blocking of potassium channels in myocytes has been suggested as the mechanism. An interaction with cytochrome P450 CYP3A4 inhibitor drugs like macrolide and azole antifungals is also thought to be a possible mechanism. Since mosapride has characteristics similar to cisapride, we examined the effects of mosapride on the electrocardiogram and pharmacokinetics before and after its coadministration with erythromycin. Ten healthy male volunteers were repeatedly administered mosapride 15 mg/day for 7 days followed by coadministration with erythromycin 1200 mg/day for 7 days. Coad- ministration with erythromycin resulted in a 1.6-fold increase in the Cmax of mosapride and prolongation of t1/2 from 1.6 to 2.4 hours, indicating the inhibition of mosapride metabolism. However, there were no significant differences in the QT interval and QTc between mosapride alone and concomitant use with erythromycin. There was no correlation between the electrocardiographic parameters and plasma mosapride concentrations, and no case exceeded the upper limit of the normal range of QTc. Although there was a certain pharmacokinetic interaction between mosapride and erythromycin, their coadministration did not affect the electrocardiogram at all, indicating a reduced likelihood of severe clinical adverse events like QT prolongation and torsades de pointes. (Jpn Heart J 2003; 44: 225-234)

Key words: Mosapride, Erythromycin, Drug interaction

IN recent years, cardiovascular adverse effects including QT prolongation and torsades de pointes have been documented1,2) in patients administered cisapride, a gastroprokinetic drug with serotonin 5-HT4 receptor agonistic activity. Its blocking action on delayed rectifier potassium channels in myocytes has been suggested as the mechanism.3,4) Cisapride, which is metabolized by cytochrome P450 CYP3A4 species, undergoes inhibition of its biotransformation by CYP3A4

From 1 The First Department of Internal Medicine, Nippon Medical School, Tokyo, 2 Clinical Pharmacology Center, Niiza- shiki Chuo-Sogo Hospital, Saitama, 3 Department of Pharmacology, St. Marianna Medical College, Kanagawa, 4 Department of Internal Medicine, Nippon Medical School Tama Nagayama Hopital, Tokyo, Japan. Address for correspondence: Takao Katoh, MD, First Department of Internal Medicine, Nippon Medical School, 1-1-5 Sendagi, Bunkyo-Ku, Tokyo 113-0022 Japan. Received for publication July 22, 2002. Revised and accepted August 30, 2002. 225 Jpn Heart J 226 KATOH, ET AL March 2003

Figure 1. Chemical structure of mosapride. Chemical Name: (±)-4-Amino-5-chloro-2-ethoxy-N-[[4-(4-fluo- robenzyl)-2-morpholinyl]methyl]benzamide citrate dihydrate.

inhibitor drugs like macrolide antibiotics and azole antifungals, resulting in an increase in its plasma concentration when coadministered with the inhibitor. Thus, it has been pointed out that coadministration may lead to cardiac adverse events such as QT prolongation and consequent torsades de pointes.1,2) In contrast, another novel gastroprokinetic agent, mosapride (Figure 1), which has similar pharmacological properties to cisapride,5) is reported in animal experiments to have only slight blocking activity of myocardial potassium channels compared with cisapride.4) However, the major metabolic enzyme of mosapride was shown to be CYP3A46) and therefore its plasma concentration can be elevated by its coadministration with CYP3A4 inhibitors such as those men- tioned above. Although the blocking potential of mosapride on cardiac potassium channels is quite weak, no detailed clinical studies have yet been carried out. Therefore, we conducted the present prospective study which examined the effects of mosapride on the electrocardiogram and pharmacokinetics before and after its coadministration with erythromycin in healthy adult male volunteers.

METHODS Subjects: The subjects were 10 healthy young male Japanese volunteers who had been selected based on a medical check-up prior to the study. Mean age was 23.2 ± 3.04 years. Written informed consent was obtained from all subjects. Test drugs and administration: White film-coated tablets containing 5 mg of mosapride citrate anhydride/tablet and whitish film tablets containing erythromycin stearate (100 mg titer/tablet) were used as test drugs. One tablet of 5 mg mosapride was administered t.i.d. before meals everyday for 14 consecutive days. From the 8th day, three tablets of 100 mg erythromycin were concomitantly administered q.i.d. The times of administration were 9:00 a.m., 1:00 p.m. and 6:00 p.m. for mosapride and 9:00 a.m., 1:00 p.m., 6:00 p.m. and 10:00 p.m. for Vo l 4 4 No 2 INTERACTION OF MOSAPRIDE AND ERYTHROMYCIN 227

Figure 2. Study design. Time of administration: Mosapride; 9:00 a.m., 1:00 p.m. and 6:00 p.m. Erythromycin; 9:00 a.m., 1:00 p.m., 6:00 p.m. and 10:00 p.m. Inquiry and vital signs: ∆; prior to the study, on the 15th study day, and after the study. ❍; just before and 1 hour after morning administration. ●; just before and 1, 2, 4, and 8 hours after morning mosapride administration. 12 lead ECG: ❍; just before mosapride administration. ; 1 hour after mosapride administration. ●; just before and 1, 2, and 4 hours after mosapride administration. Blood sampling for pharmacokinetics: ❍; just before and 1 hour after morning administration. ●; just before and 0.5, 1, 2, and 4 hours after morning administration.

erythromycin. These test dosages of mosapride and erythromycin were both iden- tical to those clinically prescribed. Study design (Figure 2): 1) Subjective and objective findings: The volunteers recorded their own subjective symptoms. Objective symptoms were documented at previously scheduled clinical examinations. If any abnormality appeared, the time of appearance, the symptoms, and their degree were documented. The appropriate intervention was undertaken and the study was discontinued based on either the judgment of the doctor in charge or the wishes of the volunteer. 2) Vital signs: Blood pressure and pulse rate were measured 5 minutes after rest with an automatic sphygmomanometer. 3) Electrocardiogram: A 12-lead electrocardiogram at rest was recorded just before and 1, 2, and 4 hours after drug administration in the morning on the 1st, 7,th and 14th days, and 1 hour after morning administration on the 4th, 8th, 10,th and 12th days. Holter monitoring was recorded 1 day before and on the 6th and 13th days. 4) Clinical laboratory tests: The clinical laboratory tests shown in Table I were conducted just before the study and on the 7th and 15th days. If an abnormal

Table I. Clinical Laboratory Tests

Hematology Red cell count, hemoglobin, hematocrit, white cell count, differen- tial white cell count, platelets

Blood biochemistry Total protein, albumin, urea-N, creatinine, total cholesterol, total bilirubin, ALP, GOT, GPT, γ-GTP, Mg, Ca, Na, K, Cl

Urinalysis pH, glucose, protein, occult blood, urobilinogen Jpn Heart J 228 KATOH, ET AL March 2003 value was observed, the subject was followed for at least 1 month or until the test value had returned to baseline. Pharmacokinetics: 1) Measurement of plasma concentrations: The plasma concentrations of unchanged mosapride and the metabolite M-1 were measured by high performance liquid chromatography by Shimadzu Techno-Research Inc. Japan. The plasma concentration of erythromycin was measured using an anti- bacterial bioassay by Dainippon Pharmaceutical Co., Ltd. Japan. 2) Pharmacokinetic analysis: The maximal plasma concentration (Cmax) and the time (Tmax) were those actually observed in each volunteer's time course of plasma concentration on the 7th (the final day point after administration of mosapride alone) and 14th (the final day point after co-administration with erythromycin) days. The elimination half-life (t1/2) was calculated using the least square method. The area under plasma concentration-time curve until 4 hours after administration (AUC0-4) was calculated by the trapezoidal rule. Analysis of electrocardiogram: 1) Twelve lead electrocardiogram at rest: Elec- trocardiograms obtained on the first (just before the study), 7th and 14th days were analyzed. After randomization by the controller, the RR and QT intervals were measured by three cardiologists at lead II recorded with a paper speed of 50 mm/ sec and their mean values were used for evaluation. The end of a T wave was defined as the point of full recovery to the isoelectric line. QTc was calculated using Bazett's equation. To examine the effect of mosapride alone, electrocardiograms conducted before and 7 days after administration were compared to determine the effect of coadministration with erythromycin on the electrocardiograms on the 7th and 14th days. 2) Holter monitoring: The presence or absence of arrhythmia and any changes in heart rate were examined by Holter monitoring. Approval by Institutional Review Board (IRB): The protocol was approved by the IRB of Niizashiki Chuo-Sogo Hospital. Statistical analysis: A paired two-tailed t test was used in the statistical analysis and the level of significance was 5%.

RESULTS Subjective symptoms and objective findings: Slightly soft stools in three subjects and transient light abdominal pain in one subject were observed during the study, however, all of these disappeared without any treatment. Vital signs (blood pressure and pulse rate) were found to be within the range of normal physiological variation. Pharmacokinetics: The time course of plasma concentrations of mosapride is shown in Figure 3. Plasma concentrations of mosapride increased after coadmin- Vo l 4 4 No 2 INTERACTION OF MOSAPRIDE AND ERYTHROMYCIN 229

Figure 3. Plasma concentrations of mosapride after single administration and concomitant administration with erythromycin. Values are means of 10 subjects±SD. ❍: Before administration or 24 hours after administration. ●: 0.5, 1, 2, and 4 hours after administration on the 7th and 14th days and 1 hour after administration on the 4th, 8th, 10th, and 12th days.

Table II. Phramacokinetic Parameters of Mosapride

Cmax Tmax t1/2 AUC0-4 (ng/mL) (hr) (hr) (ng•hr/mL)

7th day (mosapride alone) 42.1±22.6 0.5±0.0 1.6±0.4 62±27

14th day (with erythromycin) 65.7±23.8 0.5±0.0 2.4±0.5 114±35

istration with erythromycin. Table II presents the plasma pharmacokinetic parameters for mosapride on the 7th (final day point after administration of mosapride alone) and 14th (final day point after its coadministration with erythromycin) days. The Cmax of mosapride was 42.1 ng/mL on the 7th day and it increased to 65.7 ng/mL on the 14th day by concomitant administration with erythromycin. The T1/2 was prolonged from 1.6 to 2.4 hours and the AUC0-4 increased from 62 to 114 ng•hr/mL. No significant changes in the time course of metabolite M-1 were seen Jpn Heart J 230 KATOH, ET AL March 2003

th th between the 7 and 14 days: Cmax was 12.0 and 11.4 ng/mL, respectively. µ The mean Cmax of erythromycin was 1.5 g/mL. Electrocardiographic findings: 1) Twelve lead electrocardiogram: (1) Effect of administration of mosapride alone Table III presents the electrocardiographic parameters before mosapride administration on the 1st day (8:30 a.m.) and 7th day at the same time (before morning administration). No significant differences between the 1st and 7th days were seen in the RR and QT intervals or QTc. (2) Effect of coadministration with erythromycin Table IV shows the measured electrocardiographic parameters before and after administration of erythromycin on the 7th and 14th days, respectively. No significant differences were observed in the RR and QT intervals or QTc at any time point examined. In addition, the QTc value did not exceed the upper limit of its normal range (0.35-0.44)14) in any instance. (3) Relationship between electrocardiographic parameters and plasma concentration of mosapride Figure 4 shows the relationship between QTc and plasma concentration of

Table III. ECG Parameters during Mosapride Administration

1st day 7th day t test

RR interval 1.084±0.135 1.105±0.188 NS QT interval 0.389±0.020 0.389±0.019 NS QTc 0.374±0.016 0.372±0.029 NS

Means± SD (sec), n =10, t test: Paired t test between the 1st and 7th days.

Table IV. ECG Parameters during Mosapride Alone and with Erythromycin

Time from 7th day 14th day t test administration mosapride alone with erythromycin

RR interval before 1.105±0.188 1.107±0.150 NS after 1 hr 1.123±0.140 1.057±0.150 NS after 2 hr 1.068±0.137 1.109±0.184 NS after 4 hr 1.154±0.197 1.116±0.132 NS

QT interval before 0.389±0.019 0.394±0.021 NS after 1 hr 0.384±0.012 0.384±0.016 NS after 2 hr 0.387±0.013 0.390±0.023 NS after 4 hr 0.389±0.020 0.386±0.021 NS

QTc before 0.372±0.029 0.376±0.028 NS after 1 hr 0.363±0.013 0.375±0.020 NS after 2 hr 0.375±0.018 0.373±0.023 NS after 4 hr 0.364±0.026 0.366±0.010 NS

Means±SD (sec), n =10, t test: Paired t test between the 7th and 14th days Vo l 4 4 No 2 INTERACTION OF MOSAPRIDE AND ERYTHROMYCIN 231

Figure 4. Correlations of QTc and plasma concentration of mosapride. Points were values at 1 (◆), 2 (●) and 4 (■) hours after administration on the 7th day and at 1 ( ), 2 ( ) and 4 ( ) hours on the 14th day.

mosapride 1, 2, and 4 hours after administration on the 7th and 14th days. The correlation coefficient was 0.177 (coefficient of determination = 0.031), indicating an insignificant correlation. Further, the correlation coefficient between mosapride concentration and the RR or QT interval was -0.215 (0.046) or -0.156 (0.024), respectively. Thus, no significant relationships were observed between the electrocardiographic parameters and plasma concentration of mosapride. 2) Holter monitoring: Holter monitoring conducted the day before the study and on the 6th and 13th days of the study revealed no occurrence of clinically notable arrhythmia or ST-T changes in any subject. Clinical laboratory test: Three volunteers exhibited slightly high total bilirubin values in the examination before the study, however, they were judged to pose no clinical problem so the subjects were enrolled. The value was unchanged in two throughout the study and decreased to the normal range in one on the 7th day. Jpn Heart J 232 KATOH, ET AL March 2003

DISCUSSION Mosapride was synthesized and developed by Dainippon Pharmaceutical Co., Ltd. of Japan as a novel gastroprokinetic agent possessing highly selective serotonin 5-HT4 receptor agonistic activity but without a dopamine D2 antagonis- tic effect.5) Clinical use of bethanechol and metoclopramide in the 1960s marked the beginning of development of gastroprokinetic agents. Thereafter, with advances in the understanding of the mechanisms to control gastrointestinal motility, acla- tonium napadisilate, domperidone, trimebutine maleate, and cisapride were developed in the 1980s and itopride in the 1990s. These agents have been shown to remove the equivocal symptoms accompanying chronic gastritis by improving the decreased gastrointestinal motility. However, it has been reported that metoclopramide, domperidone, and itopride exhibit adverse effects such as extrapyra- midal symptoms, lactation, and gynecomastia, while cisapride causes headache and ventricular arrhythmia. Recently, reports concerning the cardiovascular adverse effects of cisapride, such as QT prolongation and ventricular arrhythmia, including torsade de pointes, have been appearing,1,2) which is leading to the contraindication of its combination with macrolide antibacterials, azole antifungals, and other drugs known to cause QT prolongation such as terfenadine and astemizol. These con- traindicated agents are reported to be metabolized by cytochrome P450 CYP3A4 species3,6) and, moreover, the antibiotics and antifungals inhibit the CYP3A4 enzyme to cause an elevation in plasma concentration of cisapride when coadministered. It is well known that ventricular arrhythmia caused by prolongation of the QT interval is attributable to prolongation of the action potential duration and early afterdepolarization due to potassium channel blocking in ventricular myocardium.7) Comparative studies of cisapride and mosapride on action potentials in isolated papillary muscle cells of guinea pigs showed that cisapride, even at low concentrations, prolonged repolarization of the action potential while mosapride had no effect.8) In addition, cisapride also prolonged the action potential duration in isolated rabbit Purkinje fibers while mosapride had no effect.4) The potassium channel blocking potential of mosapride is reported to be around 1/400 that of cisapride.9) Cisapride caused prolongation of the QT interval and QTc in anesthetized rats and in conscious cats, whereas mosapride had no effect.8) With regards to the pharmacokinetics of mosapride, since its major metabolic enzyme is CYP3A4,6) it was thought that its plasma concentration could be elevated when it was concomitantly administered with macloride antibacterials or azole antifungals. In the present study, after 15 mg/day (t.i.d.) of mosapride was administered to healthy male volunteers for consecutive 7 days, erythromycin Vo l 4 4 No 2 INTERACTION OF MOSAPRIDE AND ERYTHROMYCIN 233

1,200 mg/day (q.i.d.) was concomitantly given for a further 7 days to determine if there were any pharmacokinetic changes. The concentration of unchanged mosapride was found to be elevated by the concomitant administration: Cmax of 42.1 ng/mL after the last administration of mosapride alone at the 7th day increased by 1.6-fold to 65.7 ng/mL after the last concomitant administration at 14 days and the t1/2 of 1.6 hours increased to 2.4 hours. These facts suggest that the metabolism of mosapride is inhibited by erythromycin. Concentrations of the metabolite M-1 were only slightly changed. It is known that external and internal factors can affect an electrocardiogram and that there is a circadian rhythm in RR and QT intervals.10,11) Therefore, in the present study we decided to compare the electrocardiographic parameters at iden- tical times (1, 2, and 4 hours after administration of mosapride) on the 7th (mosapride alone) and 14th (with erythromycin) days. No significant changes in the RR and QT intervals or in QTc were observed at any time point. Further, no significant differences were observed in the RR and QT intervals or QTc between electrocardiograms conducted before the study and on the 7th day. Also, there was no significant correlation between plasma concentration and QTc. Thus, it is con- ceivable that not only administration of mosapride alone, but also its concomitant administration with erythromycin, which leads to an elevation of the plasma concentration of mosapride, does not result in significant changes in the electrocardiographic findings. The above results suggest that mosapride itself has no effect on the electrocardiogram when used at clinical dosages and that adverse cardiovascular effects like QT prolongation and ventricular arrhythmia are only slight when mosapride is coadministered with macrolide antibacterials and azole antifungals.

REFERENCES

1. Wysowski DK, Bacsanyi J. Cisapride and fatal arrhythmia. New Engl J Med 1996; 335: 290-1. 2. Gray VS. Syncopal episodes associated with cisapride and concurrent drugs. Ann Pharmacother 1998; 32: 648- 51. 3. Piruette RK. Torsades de pointes induced by cisapride/ clarithromycin interaction. Ann Pharmacother 1999; 33: 22-6. 4. Carlsson L, Amos GJ, Andersson B, Drews L, Duker G, Wadstedt G. Electrophysiogical characterization of the prokinetic agents cisapride and mosapride in vivo and in vitro; Implications for proarrhythmic potential? J Phar- macol Exp Ther 1997; 282: 220-7. 5. Yoshida N, Ito T. Mosapride citrate (AS-4370), a new gastroprokinetic agent, is a partial 5-HT4 receptor agonist in the gut. Neurogastroenterol Motil 1994; 6: 197-204. 6. Hashizume T. Identification of human metabolic enzyme P 450 relating the metabolism of mosapride citrate [in Japanese]. Bulletin of Dainippon Pharmaceutical Co. Ltd. 1989. 7. Roden DM. Current status of class III antiarrhythmic drug therapy. Am J Cardiol 1993; 72: 44B-9B. 8. Nakatsuji K, Nose I, Yabuuchi M, Ito T. Effects of serotonin 5-HT4 agonists on the electrocardiogram of rat and cat [in Japanese]. Jap J Pharmacol 1996; 108: 99. Jpn Heart J 234 KATOH, ET AL March 2003

9. Kii Y, Ito T. Effects of 5-HT4-receptor agonists, cisapride, mosapride citrate, and zacopride, on cardiac action potentials in guinea pig isolated papillary muscles. J. Cardiovasc Pharmacol 1997; 29: 670-5. 10. Morganroth J, Brozovich FV, McDonald JT, Jacobs RA. Variability of the QT measurement in healthy men, with implications for selection of an abnormal QT value to predict drug toxicity and proarrhythmia. Am J Car- diol 1991; 67: 774-6. 11. Molnar J, Zhang F, Weiss J, Ehlert FA, Rosenthal JE. Diurnal pattern of QTc interval: how long is prolonged? Possible relation to circadian triggers of cardiovascular events. J Am Coll Cardiol 1996; 27: 76-83.