Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, June 2007, p. 1912-1917, Vol. 51, No. 6
0066-4804/07/$08.00+0 doi:10.1128/AAC.01223-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Daiichi Sankyo Pharma Development, Edison, New Jersey,1 Daiichi Sankyo Pharma Development, Chiltern Place, United Kingdom2
Received 28 September 2006/ Returned for modification 29 January 2007/ Accepted 8 March 2007
|
|
|---|
|
|
|---|
Drug-induced nephrotoxicity is a clinically significant adverse effect associated with quinolone antibiotics, such as ciprofloxacin, norfloxacin, moxifloxacin, and pefloxacin (1, 14, 15, 17, 21, 22). The clinical effects of quinolone-associated renal toxicity include acute renal failure, decreased creatinine clearance (CLCR), elevations in serum creatinine (SCr) and blood urea nitrogen (BUN) levels, hematuria, albuminuria, and nephritic syndrome (13-15, 17, 22). Azotemia or an elevated SCr level occurs in 0.2% to 1.3% of patients treated with ciprofloxacin, norfloxacin, ofloxacin, or pefloxacin (21). The exact mechanism for quinolone-mediated nephrotoxicity is unclear; but both type III hypersensitivity reactions, which lead to allergic interstitial nephritis, and crystal nephropathy, which leads to direct tubular damage, were reported in patients treated with ciprofloxacin (2, 22, 23). Therefore, evaluation of renal function is an important component of clinical development for novel quinolones (30). Evaluation of the effects of DX-619 on renal function was further warranted by observations of elevated SCr levels in healthy volunteers receiving single and repeated doses >200 mg in phase I clinical studies.
CLCR is a function of both tubular secretion and glomerular filtration (6). Human organic cation transporters (hOCT2 from the SLC22 family) regulate creatinine uptake from plasma at the basolateral membrane of renal proximal tubules and the subsequent excretion of creatinine into the tubular lumen (33). Variants of SLC22 exhibit altered transport functions that may affect the renal elimination of drugs and creatinine secretion. Assessment of the glomerular filtration rate (GFR) by measurement of the clearance of the nonionic contrast medium iohexol is considered a more effective and accurate way to measure nephrotoxicity than assessment of CLCR because iohexol is excreted entirely by glomerular filtration, with no tubular secretion (11, 20). The measurement of plasma clearance of iohexol can thus distinguish the effects of a drug on the GFR from the effects on creatinine uptake. We undertook an iohexol study with healthy subjects to determine whether DX-619 reduces the GFR compared with placebo.
|
|
|---|
2 years postmenopausal or surgically sterilized or to use appropriate contraception throughout the study period if they were of childbearing potential. Pregnant or lactating women and subjects with current or previous clinically significant illnesses were not eligible. Subjects were not eligible if they had ever received DX-619 or if they had taken prescribed topical or systemic medications or herbal supplements within 14 days of dosing or over-the-counter topical or systemic medications within 7 days of dosing. Exclusion criteria included a history of smoking, drug or alcohol abuse, photosensitivity to any medications, blood donation within 56 days, or any allergies. All subjects provided written informed consent before screening, and the study protocol was approved by an independent institutional review board. The study was conducted at DaVita Clinical Research Unit (Minneapolis, MN) according to good clinical practices and the guidelines of the Declaration of Helsinki. Study design. This was a blinded placebo-controlled parallel-design study with healthy subjects randomized to receive either DX-619 at 800 mg or placebo (saline) as a 1-h intravenous (i.v.) infusion (volume, 400 ml) given once daily for 4 days via a peripherally inserted central catheter. The subjects were admitted to the clinical facility 2 days before dosing and remained confined until day 6 of the study; they then returned for follow-up assessments on days 7, 9, 11, 12, 15, 18, 21, and 24. Blood samples (5 ml) were collected for analysis of the DX-619 concentration in plasma before dosing on days 1 and 4; at the end of the infusion and 1, 3, and 9 h after the infusion on days 1 and 4; and 24 h after the infusion on day 5. Blood samples (5 ml) were collected for determination of the DX-619 concentration in plasma ultrafiltrates on days 1 and 4 predosing and 1 and 4 h after the start of the DX-619 infusion. Urine for analysis of the DX-619 concentration was collected for 24 h on day 4. Adverse events were assessed daily on days 1 to 7 and at each visit on days 9, 11, 12, 15, 18, 21, and 24; vital signs were assessed (supine blood pressure and pulse, oral temperature) on days 1 to 6, 11, and 12; routine laboratory safety tests (chemistry panel, complete blood cell count, urinalysis) were performed on days 5 and 12; and triplicate 12-lead ECGs were performed on days 1 to 4.
Measurement of GFR. A single bolus infusion of 20 ml iohexol (Omnipaque; 300 mg/ml; Amersham, Princeton, NJ) was administered on days 1, 4, and 11 at the time of day corresponding to 1 h after the end of DX-619 infusion on day 4. A 10-ml blood sample was collected at 120, 150, 180, 210, 240, 270, and 300 min after iohexol dosing for analysis of the iohexol concentration by using a validated bioanalytical method (high-performance liquid chromatography with UV detection; Medeval, London, United Kingdom) (11, 20). The GFR, assessed by measurement of the rate of clearance of iohexol from plasma (CL0), was calculated by the equation CL0 = dose x (b/c1) (ml/min), where dose is the amount of iohexol, b is the elimination rate constant of iohexol, and c1 is the intercept on y axis where b and c1 were calculated by linear regression of the iohexol plasma concentration-time profile. The clearance (CL) of iohexol was corrected for nonimmediate mixing of iohexol by the equation CL = (0.990778 x CL0) [0.001218 x (CL0)2]. The GFR was normalized to a body surface area (BSA) of 1.73 m2.
Measurement of SCr and BUN levels. Blood samples (5 ml) for analysis of SCr and BUN levels were obtained prior to the start of the DX-619 infusion and at the corresponding time of day on days 2 to 9, 11, 12, 15, 18, 21, and 24. Samples for analysis of SCr and BUN were also collected 10 h after the predose sampling on days 1, 1, 4, and 11 and 4 h after predose sampling on day 1.
Measurement of CLCR in 24-h urine. Urine for analysis of the creatinine concentration was collected over 24-h intervals at the baseline (24 to 0 h), after the end of treatment (day 4, 72 to 96 h), and on day 11 (240 to 264 h). The SCr concentration used in this calculation was the value measured at the start of the 24-h urine collection. CLCR was calculated by the formula CLCR = ([amount of creatinine in urine])/[(SCr concentration·24 h)·(1.73/BSA)], where [amount of creatinine in urine] is the concentration of creatinine in urine·total volume urine in 24 h.
Pharmacogenetics. A separate consent was obtained from each patient for participation in the pharmacogenetic portion of the study. A 6-ml whole-blood sample was shipped to EPIDAUROS Biotechnologie (Bernried, Germany) for extraction and purification of genomic DNA. DNA fragments of the SLC22A2 gene were amplified by PCR to identify variants (134InsA, C160T, T481C, A493G, G495A, G808T, C890G, C1198T, or A1294C).
PK analyses. Plasma, plasma ultrafiltrate, and urine samples were analyzed for DX-619 by a validated, sensitive high-performance liquid chromatography methods with tandem mass spectrometric detection developed by PPD Inc, (Richmond, VA). The assay was linear over the standard curve range of 0.02 to 10 µg/ml for DX-619 in plasma, 0.01 to 5.0 µg/ml for DX-619 in plasma ultrafiltrate, and 0.1 to 50.0 µg/ml for DX-619. The validation of the assay allowed extension of the upper plasma concentration limits to 40 µg/ml of DX-619 by a fourfold dilution with control human plasma. For DX-619 in plasma ultrafiltrate, the between-run and within-run precisions (expressed as the coefficient of variation, in percent) for quality control standards (25, 250, and 4,000 ng/ml) ranged from 1.82% to 4.67%, and accuracy (expressed as analytical recovery, in percent) ranged from 100.81% to 96.02%. For total DX-619 in plasma, the between-run and within-run precisions for the quality control standards (50, 500, and 8,000 ng/ml) ranged from 2.91% to 8.95%, and accuracy ranged from 97.12% to 99.16%. For DX-619 in urine, the between-run and within-run precision for quality control standards (250, 2,500, and 40,000 ng/ml) ranged from 2.80% to 9.57%, and accuracy ranged from 96.4% to 101.86%.
PK parameters were estimated by noncompartmental analysis of the plasma concentration-time data or cumulative urine data by using the SAS program (version 8.2; SAS Institute, Inc., Cary, NC). The parameters for plasma included the area under the plasma concentration-time curve from time zero to the time of the last measurable concentration (AUCt) by using the linear (ascending)/log (descending) rule, the maximum concentration in plasma (Cmax) by observation of the data without interpolation, and the time to Cmax (Tmax). The parameters calculated on day 4 included the steady-state area under the curve (AUCss,24), measured during the 25-h period from the beginning of the infusion on day 4 to 24 h after the infusion by using the linear (ascending)/log (descending) rule; Cmax and Tmax after the fourth dose, which corresponded to concentrations at steady-state (Css,max and Tss,max, respectively); and CL, where CL = dose/(AUCss,24). PK parameters for urinary excretion of DX-619 determined on day 4 were the cumulative amount excreted in urine during the 24-h period [Xu(0 to 24)]; the fraction excreted in urine in 24 h [Fe(0 to 24); in percent], calculated as Xu(0 to 24)/dose·100; and renal clearance (CLR), where CLR = Xu(0 to 24)/AUCss,24. The level of protein binding of DX-619 was calculated at each plasma ultrafiltrate sampling time point from the free concentration in the plasma ultrafiltrate and the total concentration in the corresponding plasma sample.
Data analysis. Descriptive statistics were calculated for demographic data, with tabulation of categorical measures and the frequency and percentage of subjects in the per protocol (PP) population, which included all subjects who were randomized and treated for 4 days with no major protocol deviations and with GFR assessment by iohexol analysis at day 4. The effects of DX-619 on renal function were evaluated in the PP population. The GFR changes from the baseline were assessed at days 4 and 11 by a noninferiority mixed-effects method for comparison of the lower limit of a 95% confidence interval (CI) for the difference in the mean change in the GFR from the baseline between treatment groups. The prespecified noninferiority margin for the clinically significant difference in CI between treatment groups on days 4 and 11 was 12 ml/min/1.73 m2. The 95% CIs were estimated from a mixed-models analysis of the data from the PP population and included fixed effects for baseline, treatment groups, time, and treatment-time interaction with the addition of GFR data from day 11 and a random effect for subject. The model was fit with a compound symmetry or an unstructured variance structure, with acceptance of the model determined with the largest Akaike information criterion (5). Residual analysis was used to test the fit of the model with noninferiority assessed by the use of nonparametric 95% CIs, calculated by a Mann-Whitney test for poorly fit models (16). DX-619 was noninferior to placebo with respect to GFR if the lower limit of the CI of the difference between DX-619 and placebo was not less than 12 ml/min/1.73 m2. The effects of DX-619 treatment on the changes from baseline in CLCR and SCr were determined for the PP population by using mixed-effects models. These mixed-effects models included fixed effects for baseline, treatment group, visit, and treatment group-visit interaction, with a random effect for subject. The variance structure and residual analysis for model validation were as described above for the GFR.
|
|
|---|
Demographics. The demographics at the baseline for the PP population were comparable between subjects receiving DX-619 and subjects receiving placebo (8 women and 9 men [15 white subjects] receiving placebo and 9 women and 7 men [14 white subjects] receiving DX-619; mean age ± standard deviation, 22.8 ± 3.7 years for the group receiving placebo and 23.3 ± 4.2 years for the group receiving DX-619). The subjects in both groups had similar heights and weights and, therefore, similar BSAs. There were no differences in mean SCr or BUN levels or CLCR at the baseline between the treatment groups.
GFR. There were no differences in the mean GFR on day 4 or 11 between the subjects receiving DX-619 and the subjects receiving placebo (Table 1). The two-sided 95% CI for the difference in the change in the GFR from the baseline in the DX-619 and placebo groups was above the lower limit of the prespecified noninferiority margin on days 4 and 11.
|
View this table: [in a new window] |
TABLE 1. GFR and change from baselinea
|
|
View this table: [in a new window] |
TABLE 2. Renal function parameters and change from the baselinea
|
![]() View larger version (14K): [in a new window] |
FIG. 1. Mean SCr level (95% CI) in subjects treated with DX-619 () or placebo ( ).
|
![]() View larger version (13K): [in a new window] |
FIG. 2. Mean CLCR (95% CI) in subjects treated with DX-619 () or placebo ( ).
|
PKs. The mean Cmaxs of DX-619 in plasma were 8.21 and 8.85 µg/ml on days 1 and 4, respectively, with a Tmax of approximately 1.25 h. Plasma DX-619 concentrations over time showed a moderate coefficient of variation between subjects (20 to 35%) at all time points. The mean extent of DX-619 binding to plasma proteins was 65% to 71% at the sampling time points, indicating no differences with either time course or drug concentration. On average, 27% of the DX-619 dose was excreted unchanged in the urine on day 4 (mean CL and CLR, of 9.97 liters/h and 2.6 liters/h, respectively).
General safety. Safety was determined for 41 randomized subjects who received at least one dose of the study drug. There were no serious adverse events or deaths in this study. A total of 221 and 11 adverse events were reported in the DX-619 and placebo groups, respectively. Most adverse events in the DX-619 group (198 of 221) were considered by the investigator to be treatment related; the most common of these were mild transient visual disturbances (auras and bright flashes) and mild to moderate i.v. injection site reactions. There were no significant changes in laboratory parameters, vital signs, ECGs, or findings on physical examination.
|
|
|---|
On days 4 and 11, the mean GFR normalized for BSA was comparable between the DX-619 and the placebo groups, and the corresponding mean changes in the GFR from the baseline were negligible in both groups. The lower limit of the 95% CI for the changes in GFR from the baseline for DX-619 did not exceed the prespecified clinically significant margin, which confirms that DX-619 is noninferior to placebo for a directly measured effect on the GFR. Even though direct assessment of the GFR by measurement of iohexol CL is a more accurate parameter, the SCr level and CLCR are commonly used in clinical practice to estimate renal function (11, 20). DX-619 treatment caused a moderate (30 to 40%) increase in the SCr level from the baseline, which corresponds to a decrease in CLCR from the baseline. Both increases fully reversed within 7 days after the end of treatment. There were no meaningful changes in these parameters in the placebo group. All subjects with increased SCr levels and decreased CLCR had normal GFRs. The mechanism by which DX-619 increases the SCr level appears to be inhibition of renal tubular transporters. SCr is both filtered by glomeruli and secreted by proximal tubules (3, 30). The tubular secretion of creatinine is mediated by the basolateral organic cation transporter SLC22 (33). Some medications, such as probenecid, cimetidine, and trimethoprim-sulfamethoxazole, can alter CLCR through the impairment of tubular secretion (8, 12, 18, 19, 25). We attempted to determine the SLC22 genotype in our study to better understand the differences in CLCR between treatments, but the data available were insufficient for conclusive analysis. In nonclinical in vivo studies with the rat, DX-619 was shown to inhibit the uptake of creatinine in rats, with the inhibition mediated by the organic cation transporter SLC22 (27). In the same study, the magnitude of the increase in the SCr level after cotreatments with DX-619 and cimetidine or probenecid was comparable to that after treatment with DX-619 alone, with no changes in renal histopathology (27). Therefore, it is possible that DX-619 inhibited the uptake of SCr from the renal circulation into the renal tubular epithelium, which results in increased SCr levels. The inhibition of tubular active transport of creatinine is a reversible and saturable process that would correspond to our findings of the complete reversibility of the increase in SCr levels after the cessation of dosing with DX-619, the moderate degree of increase, and the corresponding normal GFR. The abnormal SCr levels would thus not be a clinically significant safety issue for the use of DX-619 in clinical practice.
In conclusion, DX-619 does not impair the GFR after repeated daily administration of 800 mg by i.v. infusion. The moderate increase in SCr levels and the decrease in CLCR in subjects receiving DX-619 are fully reversible and are probably caused by the inhibition of the hOCT2-mediated tubular secretion of creatinine. Therefore, DX-619 is not expected to cause nephrotoxicity in patients with infectious diseases. The normal GFR results indicate that the potential for renal toxicity from treatment with DX-619 may be differentiated from that from treatment with other quinolones, such as ciprofloxacin, norfloxacin, moxifloxacin, and pefloxacin (1, 14, 15, 17, 21, 22), and from treatment with other antibiotics, such as aminoglycosides, imipenem-cilastatin, and vancomycin (26, 28, 29, 32). Nevertheless, renal function monitoring is warranted for patients with infectious diseases receiving DX-619, particularly those who have underlying renal insufficiency.
We are grateful to D. Craig Brater, Indiana University School of Medicine, Indianapolis, for valuable input into the study design and to Prism Communications LLC of Morristown, NJ, for technical assistance in preparation of the manuscript.
Published ahead of print on 19 March 2007. ![]()
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»