| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
DRUG INTERACTIONS |
From the University of North Carolina at Chapel Hill, School of Pharmacy and School of Medicine, Chapel Hill, North Carolina.
Address for reprints: Robert E. Dupuis, PharmD, University of North Carolina at Chapel Hill, Division of Pharmacotherapy, Kerr Hall CB#7360, Chapel Hill, NC 27599-7360.
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Key Words: Norfloxacin • metronidazole • mycophenolate mofetil • drug interactions
MMF is rapidly and presystemically hydrolyzed to MPA with a 94% bioavailability.3 The parent compound, MPA, is then metabolized to an inactive, stable phenolic glucuronide (mycophenolic acid glucuronide [MPAG]), which is excreted into the bile and urine of healthy subjects and represents about 70% of the dose in urine.2 There is some evidence of an acyl glucuronide conjugate in the plasma of patients with impaired renal function, although this is a minor metabolite based on urine recovery.4,5 Following excretion of MPAG into the bile, it is believed to undergo enterohepatic recirculation (EHC). EHC is suggested by the presence of a second MPA peak concentration at approximately 6 to 8 hours after dosing and the recovery of MPAG in the bile of orthotopic liver transplant patients.3,6 EHC is initiated by ß-glucuronidase, which cleaves glucuronide conjugates in the intestine releasing the parent compound, making it available for reabsorption. This enzyme is produced by gram-negative aerobic and anaerobic bacteria, which are part of the normal human intestinal flora.7
Because EHC is mediated through endogenous bacteria in the intestine, it is possible that reducing bacterial content with broad-spectrum antibiotic therapy may result in a reduction of the recirculation of MPA. Similar interactions have been documented with oral contraceptives, such as the reduction in the area under the curve (AUC) of ethinylestradiol with concomitant antibiotic therapy.8 A study by Schmidt et al9 reported a substantial reduction in the AUC of MPA in a study of liver transplant recipients when bowel decontamination was employed, relative to MPA exposure at 4 to 8 days after antibiotic treatment.
MPA exhibits an immunosuppressive effect, which is highly dependent on systemic exposure, as measured by the AUC. Hale and colleagues10 established that a lower AUC of MPA resulted in an increased risk of rejection in kidney transplant recipients. Because the EHC of MPA has been reported to contribute, on average, to 40% of the total AUC exposure,1 concomitant antibiotic treatment might be expected to significantly reduce the EHC of MPA. This could lead to suboptimal drug exposure (AUC) and subsequent clinical failure and to organ rejection. The objective of this study was to evaluate the potential drug interaction of MMF in combination with antibiotics in healthy subjects.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Blood samples were collected prior to study initiation and preceding each treatment period and were used to evaluate complete blood count, electrolyte, and liver function tests and pregnancy status. Any laboratory value obtained outside the normal range or pregnancy resulted in exclusion from the study. Subjects were monitored for 1 week postdose in each treatment period for any medication-related adverse events.
The study protocol was approved by the institutional review board of the University of North Carolina at Chapel Hill. All participants gave written informed consent before the initiation of the study.
Study Protocol
The study was divided into 4 treatment periods. A baseline treatment period established the pharmacokinetic profile for a single 1-g oral dose of MMF (Cellcept; Roche Laboratories, Nutley, NJ) given alone. Following this baseline period, subjects were randomized into 1 of 6 possible treatment sequences for the remaining 3 antibiotic treatment periods. Norfloxacin (NOR; Noroxin; Roberts Pharmaceuticals, Eatontown, NJ), 400 mg bid; metronidazole (MET; Schein Pharmaceuticals, Florham Park, NJ), 500 mg tid; or the combination of norfloxacin and metronidazole (NOR + MET) comprised the 3 antibiotic regimens. The antibiotic treatment periods were each 5 days in length and were separated by a 1-week washout. Antibiotics were administered 3 days prior to administration of the single 1-g oral dose of MMF, on the day of the MMF dose, and on the day following the MMF dose. Subjects fasted overnight preceding each MMF dose. Antibiotics were administered 2 hours prior to MMF during the 3 antibiotic periods. Blood samples (7 mL) were obtained at 0, 0.5, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 24, 36, and 48 hours after the MMF dose in each period. Urine was collected over 0- to 4-, 4- to 8-, 8- to 12-, 12- to 24-, 24- to 36-, and 36- to 48-hour intervals following MMF administration. Samples were obtained in heparin-containing Vacutainers. Plasma was centrifuged at 2500 rpm for 10 minutes at 4 °C. Plasma and urine samples were aliquotted and stored at –20 °C until analysis.
Sample Analysis
Plasma samples were analyzed for the presence of MPA and MPAG (expressed as MPA molar equivalents) using reversed-phase high-pressure liquid chromatography (HPLC). Suprofen, the internal standard, was added to 0.25 mL of plasma. Acetonitrile 0.75 mL was added for protein precipitation. The samples were then centrifuged at 3000 rpm, and the supernatant was decanted and evaporated under nitrogen stream. The residue was reconstituted with 25% methanol/0.1% trifluoroacetic acid. Urine preparation was simply a 20-fold dilution with blank urine prior to injection.
The mobile phase consisted of 52% methanol/0.1% trifluoroacetic acid at 1.5 mL/min. An Axxiom C18 (150 x 4.6 mm id, 5 µm; Moorpark, Calif) column coupled to an RP-18 guard column (15 x 3.2 mm id, 7 µm) was used. Samples were assayed using UV detection at 250 nm. An HP1100 running Chemstation A.05.01 (Hewlett Packard) software was used for recording and storing the data.
The calibration curve in plasma was linear over the detectable range up to 50 µg/mL for MPA and MPAG based on the peak area ratio of MPA or MPAG relative to suprofen. The lower limit of detection (LOD) was 0.08 µg/mL for MPA and 0.2 µg/mL for MPAG in plasma. For MPA, this sensitivity represented about 0.5% of Cmax concentrations. In urine, the LOD was 1 µg/mL for MPA and 5 µg/mL for MPAG, with linearity up to 40 µg/mL for MPA and 2000 µg/mL for MPAG.11 In plasma, the coefficient of variation (CV) for MPA and MPAG ranged from 2.4% to 9.3% and 5.9% to 11% for intraday and interday variabilities, respectively. In urine, the CVs for MPA and MPAG were 1.3% to 6.0% and 1.4% to 4.3% for intraday and interday variabilities, respectively.
MPA and suprofen were obtained from Sigma Co (St. Louis, Mo), authentic MPAG was synthesized from MPA,11 and HPLC solvents were obtained from Mallinckrodt (St. Louis, Mo).
Pharmacokinetic Analysis
Pharmacokinetic parameters were determined for both MPA and MPAG using WinNonlin version 3.1 (Pharsight Corporation, Cary, NC). Maximum concentration (Cmax) and time to peak (tmax) for both parent and glucuronide were directly obtained from the plasma concentration-time curve. Area under the plasma concentration-time curve (AUC0-48) was calculated using the trapezoidal method. AUC0-
values were also calculated by adding the value of area extrapolated to infinity. Oral clearance for MPA was calculated by dividing the dose by the AUC0-48 or to the last detectable time point because the extrapolated AUC48- was variable and difficult to estimate due to apparent enterohepatic recycling but, in general, was a small percentage of the total AUC. Renal clearance for MPAG was calculated by dividing the amount excreted in urine over 48 hours, Ae0-48, by the AUC0-48. The fraction excreted into urine was calculated by dividing Ae0-48 by the dose in MPA molar equivalents, assuming F =1.
Statistical Analysis
Statistical analysis of pharmacokinetic parameters was performed using SAS version 6.12 (SAS Institute, Cary, NC). A repeated-measures MIXED model analysis was conducted with a post hoc Scheffe test applied for multiple comparisons. A P 
.05 was considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
A representative log-linear plot of plasma MPA concentration versus time is illustrated in Figure 1, which shows visible reductions in the secondary peak of MPA for each of the antibiotic treatment periods. Mean (± SD) pharmacokinetic parameters for MPA and MPAG are summarized in Table I. Cmax remained constant across the antibiotic treatment periods for both MPA and MPAG, with no significant differences seen in the Cmax of MPA or MPAG. The tmax values for MPA ranged between 0.5 and 3 hours, whereas tmax for MPAG ranged from 1 to 3 hours. Of note, due to the impreciseness of estimates of AUC48-, we applied to the pharmacokinetic analysis AUC0-48 or to the last detectable time point, rather than AUC0-. This impreciseness of estimating AUCt,last- appears to be due to extensive recycling of the parent and metabolite, causing random increases in concentrations in the terminal phase and making the extrapolated AUC highly variable and less reliable because they relied on poor estimates of elimination rate constants. It was not possible to conduct the study with dosing to steady state; thus, using AUC0-
as prolonged immunosuppression in healthy volunteers was unacceptable. The AUC0-t,last of MPA was reduced from the baseline value by an average of 10%, 19%, and 33%, corresponding to NOR, MET, and NOR + MET, respectively (Figure 2). The AUC0-t,last of MPAG was decreased, on average, by 10%, 27%, and 41% from baseline for NOR, MET, and NOR + MET, respectively, with variable reductions between subjects (range, 5%-80%). The reduction in the AUC of MPA was statistically significant only for differences seen in NOR + MET when compared to baseline (P = .01). Reductions in the AUC of MPAG were significant for both MET and NOR + MET when compared to baseline (P =.03and P = .001, respectively), even though the ratio of MPA/MPAG remained unchanged across treatment periods. Figure 3 illustrates the individual AUC of MPA and MPAG between baseline and NOR + MET. In addition, no period or sequence effects were seen.
|
|
|
|
Table I also summarizes apparent clearance parameters (± SD) for MPA and MPAG, assuming complete bioavailability. Clearance, measured using the AUC over 48 hours or to the last measurable sampling point for MPA, showed an average increase of 25%, 45%, and 84% from baseline for NOR, MET, and NOR + MET, respectively. Renal clearances remained constant for MPAG across all treatment periods at an average value of 0.5 mL/min/kg. Renal clearance of MPA was not determined because urinary MPA concentrations were below the limit of detection of the assay in most individuals, as most material in urine was present as MPAG. In some subjects, MPA was excreted in minor, measurable amounts over the first interval (0-4 hours), but no further measurements were possible due to assay limitations. Ae0-48 for MPAG was an average of 80% of the dose (based on molar equivalents) in the baseline study and showed appreciable reductions with antibiotic treatments. Ae0-48 of MPAG was reduced by 15%, 27%, and 37% for NOR, MET, and NOR + MET, respectively, relative to baseline. These reductions were significant between baseline and MET (P = .0322) and between baseline and NOR + MET (P = .001). The percentage of dose excreted as MPAG over 48 hours decreased from 82% to 51% between baseline and NOR + MET. Fecal excretion of MPA or MPAG was not measured in this study.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Evaluation of the ratios of MPA/MPAG over the entire study resulted in substantial variability between subjects (mean CV = 40%), whereas much less variability occurred within subjects (mean CV = 17%). Due to the reproducibility of this ratio across treatment periods in individual subjects, an alteration in formation clearance of the glucuronide is not likely the source of the observed increased apparent clearance of MPA.
MPA is highly bound to albumin (>97%) in plasma.12 With the low plasma protein binding of MET at less than 20% and NOR at 15%, such an interaction is unlikely to occur.13,14 No change in the renal clearance of MPAG was observed, suggesting that any interaction was not a function of altered renal clearance of MPAG. However, there were reductions in the Ae0-48 of MPAG with the antibiotics, which may be explained by the reduced EHC with drug or metabolites increasingly eliminated in the feces.
Several other factors support the conclusion that the enhanced apparent oral clearance is likely to be a result of a reduction in EHC. The secondary increase in the plasma concentration-time profile for MPA, commonly associated with EHC, was notably reduced, although not uniformly eliminated, in the antibiotic treatment periods. Furthermore, most subjects showed a reduction in the AUC0-t,last of MPA with antibiotics. The amount of MPAG excreted in the urine also decreased with antibiotics. Due to the polar nature of the glucuronide conjugate, it would be unlikely that this metabolite would be reabsorbed intact, thereby increasing fecal elimination of MPA equivalents. By increasing the fraction of the dose excreted in the feces, the fraction excreted renally would be reduced, as was observed in this study. Feces were not collected and analyzed in this study but may have further validated the proposed mechanism of the drug-drug interaction. Mechanical issues such as diarrhea may have contributed to reduced absorption of MPA. However, the 2 subjects who experienced this adverse event during the study day when MMF was administered did not demonstrate the larger changes in MPA or MPAG AUCs. The other 2 brief episodes of loose stools occurred during the study days with antibiotics.
The effect of antibiotics seems to be an additive effect, with NOR reducing the AUC of MPA on average by 10%, MET exhibiting a 19% reduction, and the combination having an approximate sum of the two (33%). The human intestinal flora consists of a wide variety of anaerobic and gram-negative aerobic bacteria.15 These bacteria have been shown to produce the ß-glucuronidase enzyme in rats.16,17 Studies have reported the effect of NOR and MET on reducing intestinal bacterial content. Norfloxacin obliterates Enterobacteriacea within 1 week of antibiotic treatment.18 Metronidazole has significant anti-microbial activity against ß-glucuronidase producing anaerobic bacteria.19,20
The norfloxacin phase showed no statistically significant reductions in AUC. Although Enterobacteriacea produces ß-glucuronidase, it does not constitute the largest fraction of intestinal bacteria. Therefore, reductions in bacteria secondary to NOR treatment may only reduce recycling by a small proportion. This finding is consistent with a study conducted by the manufacturer of Cellcept,1 which consisted of concomitantly administered trimethoprim/sulfamethoxazole, which has similar antimicrobial activity as NOR, with MMF. The study found that there was no significant reduction in the AUC of MPA.1 The bacterial specificity of MET appears to be consistent with the observed reductions in EHC found in this study. Metronidazole demonstrated a greater reduction in the AUC of MPA when compared to MMF alone and NOR. The combination of NOR + MET exhibited the greatest reduction in MPA AUC, resulting in approximately additive reductions.
Statistical differences were observed between the AUC0-t,last of MPA and MPAG with NOR + MET when compared to baseline. The amount of MPA and MPAG excreted in the urine was statistically different between MET and NOR + MET when compared to baseline. Due to the many confounding factors, AUC and oral clearance can be more variable pharmacokinetic parameters than Ae. As such, a statistical difference might be more difficult to detect.
In evaluating individual subjects, 6 (55%) of 11 subjects had reductions in the AUC0-t,last of MPA greater than 15 µg·h/mL in both MET and NOR + MET treatment periods of this study. Furthermore, 1 of 9 and 3 of 11 subjects exhibited AUCs of MPA reduction of more than 40 µg·h/mL in MET and NOR + MET, respectively. These findings have potential clinical implications. In a study of kidney transplant recipients receiving chronic MMF therapy, a relationship between MPA AUC and the likelihood of acute rejection in kidney transplant recipients was reported. The authors determined that a change in MPA AUC of 1 µg·h/mL would result in a change of efficacy of 4%.10 Although the extrapolation of potential reductions in the AUCs of MPA in transplant patients at steady state cannot be made directly from data from healthy subjects after single doses of MMF, the data reported here suggest a potentially clinically significant drug interaction that could occur in transplant patients undergoing concomitant MMF and aggressive antianaerobic/aerobic antibiotic treatments. Moreover, this controlled and randomized study in normal volunteers verifies the previous observation reported by Schmidt et al9 in 2 of 6 patients. The previous study in liver transplant recipients receiving concurrent cyclosporine was confounded by the variable washout period between antibiotic treatment and the control phase, in which the antibiotic effect was substantial in only 2 patients who had longer washout periods of 7 and 8 days.9
Concomitant antibiotic treatment with MMF therapy in healthy subjects resulted in a reduction in the AUC of MPA, the pharmacologically active species, and this effect is likely a result of reduced EHC. The reductions appear to be a function of the spectrum of antimicrobial activity and the aggressiveness of bacterial obliteration, with antibiotics specific for anaerobes causing greater reductions in the AUC of MPA. With the recognized relationship between the AUC of MPA and the risk of organ rejection, this interaction may have clinical implications in the therapy of patients after solid organ transplant.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Submitted for publication June 28, 2004; Revised version accepted September 28, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
1. Mycophenolate mofetil [prescribing information]. Nutley, NJ: Hoffman-La Roche; 1995.
2. Allison AC, Eugui EM. Preferential suppression of lymphocyte proliferation by mycophenolic acid and predicted long-term effects of mycophenolate mofetil in transplantation. Transplant Proc. 1994; 26: 3205-3210.[Web of Science][Medline] [Order article via Infotrieve]
3. Bullingham R, Monroe S, Nicholls A, Hale M. Pharmacokinetics and bioavailability of mycophenolate mofetil in healthy subjects after single-dose oral and intravenous administration. J Clin Pharmacol. 1996;36: 315-324.[Abstract]
4. Shipkova M, Schutz E, Armstrong VW, Niedmann PD, Oellerich M, Wieland E. Determination of the acyl glucuronide metabolite of mycophenolic acid in human plasma by HPLC and EMIT. Clin Chem. 2000;46: 365-372.
5. Bullingham RES, Nicholls AJ, Kamm BR. Clinical pharmacokinetics of mycophenolate mofetil. Clin Pharmacokinet. 1998; 34: 429-453.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
6. Shaw L, Solinger HW, Halloran P, et al. Mycophenolate mofetil: a consensus panel. Ther Drug Monit. 1995;17: 513-516.
7. Sperker B, Backman JT, Kroemer H. The role of ß-glucuronidase in drug disposition and drug targeting in humans. Clin Phamacokinet. 1997;33: 18-31.
8. Joshi JV. A study of interaction of low-dose combination oral contraceptive with ampicillin and metronidazole. Contraception. 1980; 22: 643-652.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
9. Schmidt LE, Rasmussen A, Norrelykke MR, Poulsen HE, Hansen BA. The effect of selective bowel decontamination on the pharmacokinetics of mycophenolate mofetil in liver transplant patients. Liver Transpl. 2001;7: 739-742.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
10. Hale MD, Nichols AJ, Bullingham RE, et al. The pharmacokinetic-pharmacodynamic relationship for mycophenolate mofetil in renal transplantation. Clin Pharmacol Ther. 1998;64: 672-683.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
11. Wiwattanawongsa K, Heinzen EL, Kemp DC, Dupuis RE, Smith PC. Determination of mycophenolic acid and its phenol glucuronide metabolite in human plasma and urine by HPLC. J Chromatogr B. 2001;763: 35-45.[CrossRef][Medline] [Order article via Infotrieve]
12. Nowak L, Shaw L. Mycophenolic acid binding to human serum albumin: characterization and relation to pharmacodynamics. Clin Chem. 1995;41: 1011-1017.
13. Sanvordeker DR, Chien YW, Lin TK, Lambert HJ. Binding of metronidazole and its derivatives to plasma proteins: an assessment of drug binding phenomenon. J Pharm Sci. 1975;64: 1797-1803.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
14. Okezaki E, Terasaki T, Nakamura M, Nagata O, Kato H, Tsuji A. Serum protein binding of lomefloxacin, a new antimicrobial agent, and its related quinolones. J Pharm Sci. 1989;78: 504-507.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
15. Drasar BS, Shiner M, McLeod GM. Studies on the intestinal flora. Gastroenterology. 1969;56: 71-79.[Web of Science][Medline] [Order article via Infotrieve]
16. Hawksworth G, Drasar BS, Hill MJ. Intestinal bacteria and the hydrolysis of glycosidic bonds. J Med Microbiol. 1971;4: 451-459.
17. Gadelle D, Raibaud P, Sacquet E. ß-Glucuronidase activities of intestinal bacteria determined both in vitro and in vivo in gnotobiotic rats. Appl Environ Microbiol. 1985;49: 682-685.
18. Boerema JBJ, Olthof BJ, Van Saene HKF. Effects of norfloxacin on the faecal flora in patients with complicated urinary tract infections. Scand J Infect Dis Suppl. 1986;48: 27-31.[Medline] [Order article via Infotrieve]
19. Tally FP, Goldin BR, Sullivan N, Johnston J, Gorbach SL. Antimicrobial activity of metronidazole in anaerobic bacteria. Anti-microb Agents Chemother. 1978;13: 460-465.
20. Barry RE, Chow AW, Billesdon J. Role of intestinal microflora in colonic pseudo-obstruction complicating jejunoileal bypass. Gut. 1977;18: 356-359.
CiteULike 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
