AAC Accepted Manuscript Posted Online 18 May 2015 Antimicrob. Agents Chemother. doi:10.1128/AAC.00354-15 Copyright © 2015, American Society for Microbiology. All Rights Reserved.
1
1
Pharmacokinetics of Second-line Anti-Tuberculosis Drugs after Multiple
2
Administrations in Healthy Volunteers
3 4
Sang-In Park,a Jaeseong Oh,a Kyungho Jang,a Jangsoo Yoon,a Seol Ju Moon,a Jong Sun Park,b
5
Jae Ho Lee,b Junghan Song,c In-Jin Jang,a Kyung-Sang Yu,a Jae-Yong Chung,d,#
6 7
a
8
of Medicine and Hospital, Seoul, Republic of Korea
9
b
Department of Clinical Pharmacology and Therapeutics, Seoul National University College
Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Seoul
10
National University College of Medicine, Seoul National University Bundang Hospital,
11
Gyeonggi-do, Republic of Korea
12
c
13
National University Bundang Hospital, Gyeonggi-do, Republic of Korea
14
d
15
Hospital, Gyeonggi-do, Republic of Korea.
16
Running title: Second-line Anti-Tuberculosis Drugs Pharmacokinetics
17
#
18
Department of Clinical Pharmacology and Therapeutics, Seoul National University, College
19
of Medicine and Bundang Hospital, Seongnam, Gyeonggi-do, Republic of Korea
20
Phone: 82-31-787-3955
21
Fax: 82-31-787-4045
22
E-mail:
[email protected] Department of Laboratory Medicine, Seoul National University College of Medicine, Seoul
Department of Clinical Pharmacology and Therapeutics, Seoul National University Bundang
Address correspondence to: Jae-Yong Chung, MD, PhD
2
23
Abstract
24
Therapeutic drug monitoring (TDM) of second-line anti-tuberculosis (TB) drugs
25
would allow for optimal individualized dosage adjustments and improve drug safety and
26
therapeutic outcomes. To evaluate the pharmacokinetic (PK) characteristics of clinically
27
relevant, multi-drug treatment regimens and improve the viability of TDM, we conducted an
28
open-label, multiple-dosing study on 16 healthy subjects who were divided into two groups.
29
Cycloserine 250 mg, p-aminosalicylic acid (PAS) 5.28 g, prothionamide 250 mg twice daily
30
and pyrazinamide 1500 mg once daily were administered to both groups. Additionally,
31
levofloxacin 750 mg and streptomycin 1 g once daily were administered to group 1, and
32
moxifloxacin 400 mg and kanamycin 1 g once daily to group 2. Blood samples for PK
33
analysis were collected up to 24 h following the 5 days of drug administration. The PK
34
parameters, including maximum concentration (Cmax) and the area under the plasma
35
concentration-time curve (AUCτ), were evaluated. The correlation between the PK parameters
36
and the concentration at each time-point was conducted. The mean Cmax (mg/L) and AUCτ
37
(mg·h/L) of each drug were as follows: cycloserine (24.9, 242.3), PAS (65.9, 326.5),
38
prothionamide (5.3, 22.1), levofloxacin (6.6, 64.4), moxifloxacin (4.7, 54.2), streptomycin
39
(42.0, 196.7), and kanamycin (34.5, 153.5). The results indicated that sampling at 1, 2.5, and
40
6 h post-dose is needed for TDM when all seven drugs are administered concomitantly. This
41
study indicates that PK characteristics must be considered when prescribing optimal
42
treatments for patients.
43 44
Key words: multidrug-resistant, anti-tuberculosis drugs, therapeutic drug monitoring (TDM)
3
45
Introduction
46
Tuberculosis (TB) that is resistant to isoniazid and rifampicin is considered to be
47
multidrug-resistant (MDR)—regardless of whether or not it is resistant to other first-line
48
drugs. MDR TB has become a major global health concern and, according to the Global
49
Project on Anti-Tuberculosis Drug Resistance Surveillance, MDR TB accounts for more than
50
3% of new TB cases in at least one country in every one of the six World Health Organization
51
(WHO) regions (1).
52
WHO guidelines, recommend the use of at least four different drugs to treat MDR
53
TB; it is recommended that regimens include pyrazinamide along with several second-line
54
drugs—specifically, a later-generation fluoroquinolone, ethionamide or prothionamide, an
55
injectable parenteral agent (kanamycin, amikacin or capreomycin), and either cycloserine or
56
p-aminosalicylic acid (PAS) (2). However, these treatment regimens are less effective, cause
57
more frequent side effects, and have a narrower therapeutic/toxic effect ratio than first-line
58
anti-TB drugs (3). Fortunately, therapeutic drug monitoring (TDM) can be used to overcome
59
variations in patient response and improve the success of second-line therapeutics by
60
allowing individualized therapies to be designed (4).
61
When planning TDM of second-line anti-TB drugs in a clinical setting, ease-of-use
62
and availability should be considered as well as clinical and bacteriological data. For instance,
63
for anti-TB drugs such as moxifloxacin and levofloxacin, the area under the concentration
64
versus time curve from time 0 to 24 h (AUC0-24h)/minimum inhibitory concentration (MIC) is
65
one of the most important pharmacodynamic parameters along with maximum plasma
66
concentration (Cmax)/MIC (5). However, estimation of AUC requires a minimum of six to
67
seven blood samples and is therefore impractical as a routine measure (4). A more practical
68
alternative to full pharmacokinetic (PK) sampling is the use of only a few sampling time-
69
points that are strongly correlated with Cmax or AUC. Furthermore, since second-line anti-TB
4
70
drugs are administered concomitantly in most cases, the possibility of multi-drug interactions
71
should always be considered. For this reason, it would be more useful to estimate the PK
72
characteristics of second-line anti-TB drugs in the context of a therapeutically relevant
73
dosage regimen.
74 75
Objectives
76
The objective of this study is to evaluate the PK characteristics of second-line anti-
77
TB drugs in MDR TB patients with a view to applying the results to clinical practice,
78
including prescribing. In addition, to reduce the number of sampling time-points required to
79
estimate AUC or Cmax, the most reliable sampling time-points were determined.
80 81
Methods
82
Study subjects and study design
83
Healthy male volunteers who had no history of renal or hepatic impairment and who
84
passed a rigorous on-site physical examination and clinical laboratory tests, which included
85
urine screening and a 12-lead electrocardiogram, were enrolled in the study. Written informed
86
consent was obtained prior to starting the procedures described in this study. This study was
87
approved by the Institutional Review Board (B-1309/219-003) of the Seoul National
88
University Bundang Hospital located in Seongnam, Korea, and was conducted in accordance
89
with the principles of the Declaration of Helsinki and Good Clinical Practice
90
(clinicaltrials.gov registration number: NCT02128308).
91
A randomized, open-label, parallel-group, multiple-dosing study was conducted in
92
the Seoul National University Bundang Hospital Clinical Trials Centre. The patients were
93
divided into two groups so that as many second-line anti-TB drugs as possible could be
94
studied with two different dosing regimens. The two regimens were chosen to reflect
5
95
treatments that are currently used clinically. Group 1 received cycloserine, p-aminosalicylic
96
acid (PAS), and prothionamide twice daily per os (PO); pyrazinamide and levofloxacin PO
97
once daily; and streptomycin intramuscular (IM) once daily from day 1 to day 5. For group 2,
98
the same regimen was applied, except that levofloxacin and streptomycin were replaced with
99
moxifloxacin and kanamycin, respectively (Table 1).
100 101
Pharmacokinetic assessment
102
Serial blood samples were collected in heparinized tubes at 0 (i.e., pre-dose), 0.5, 1,
103
1.5, 2, 2.5, 3, 4, 6, 8, 12, and 24 h after dosing on day 5. Blood samples were centrifuged for
104
10 min at 4 ºC at approximately 1910 ×g. Then, plasma (1 mL) was aliquoted into a
105
polypropylene tube and stored at -70 ºC until PK analysis.
106
The plasma concentrations of cycloserine, PAS, prothionamide, pyrazinamide,
107
levofloxacin, moxifloxacin, streptomycin, and kanamycin were determined using ultra-
108
performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS). This
109
analytical method was previously developed for the simultaneous measurement of the blood
110
concentrations of nine second-line anti-TB drugs (6). In brief, the analytes were separated on
111
a high-strength silica (HSS) T3 column (50.0 × 2.1 mm, 1.8 μm; Waters, Watford, UK) at a
112
flow rate of 200 μL/min. The mobile phases were 10 mM ammonium formate in 0.1% formic
113
acid (mobile phase A) and acetonitrile in 0.1% formic acid (mobile phase B). Separation was
114
achieved using a gradient program. Detection was performed using MS/MS. The inter-assay
115
calibration variability was evaluated using concentrations of 5.0–100 μg/mL for streptomycin,
116
kanamycin, cycloserine, and PAS; 0.5–10 μg/mL for prothionamide; and 1.0–20 μg/mL for
117
moxifloxacin and levofloxacin. The lower limits of quantification (LLOQ), defined as the
118
lowest concentration with a precision < 20% and an accuracy within ± 20% for each drug,
119
were as follows: 5.0 μg/mL for cycloserine, 5.0 μg/mL for PAS, 0.5 μg/mL for prothionamide,
6
120
1.0 μg/mL for levofloxacin, 2.5 μg/mL for streptomycin, 0.5 μg/mL for moxifloxacin, and 2.5
121
μg/mL for kanamycin.
122 123
Pharmacokinetic data analyses
124
We used the linear-up and log-down trapezoidal method to estimate the AUC during
125
a dosing interval when the concentrations reached the steady state on day 5 of the dosing
126
administration (AUCτ). The observed values were used to estimate the steady-state Cmax for
127
each drug. The apparent clearance at steady state (CL/F) was calculated as the dose divided
128
by the AUC from time 0 to infinity at steady state (AUCinf). The terminal elimination rate
129
constant (λz) was estimated from the regression line of the log-transformed plasma
130
concentrations versus time over the terminal log-linear portion. The terminal half-life at
131
steady state (t1/2) was calculated as the natural logarithm of 2 divided by λz. A non-
132
compartmental analysis using Phoenix® V1.3 (Certara, MO, USA) was used to calculate the
133
PK parameters.
134 135
Statistical analysis
136
The PK parameters of second-line anti-TB drugs were summarized using descriptive
137
statistics. Possible differences in drug exposure were determined based on the differences in
138
Cmax and AUCτ of cycloserine, PAS, and prothionamide between group 1 and group 2, and the
139
geometric mean ratio (GMR) and its 90% confidence interval (CI) for group 2 compared with
140
group 1 were estimated. Unpaired t-tests were used to compare the differences in CL/F
141
between group 1 and group 2. Analysis was performed using SAS software version 9.3 (SAS
142
Institute Inc., Cary, NC, USA). Statistical significance was confirmed if a p-value was less
143
than 0.05.
144
7
145
Selection of sampling time-points for the limited-sampling TDM strategy
146
After obtaining full PK profiles, the most reliable sampling time-point for each
147
second-line anti-TB drug was determined by correlation analyses between Cmax or AUCτ and
148
the concentrations of the drugs measured in the samples collected at each time-point after
149
dosing on day 5. This Pearson’s correlation analysis was repeated for each sampling time-
150
points from 0 to 24 h for each of the studied second-line anti-TB drugs, using R version 3.1.0
151
(R Foundation for Statistical Computing, Vienna, Austria). The optimal sampling time-points
152
were determined based on the adjusted r2 and p-value using the correlation analysis.
153 154
Results
155
Demographic characteristics
156
A total of 16 subjects were enrolled and completed the study; seven subjects for
157
group 1, and nine subjects for group 2. As the purpose of this study was to explore and
158
describe the PK characteristics of second-line anti-TB drugs, it was desirable to set the
159
minimum number of subjects based on experience from previous studies rather than on the
160
calculation required for a hypothesis study. As such, the required number of subjects for this
161
study was determined to be at least six for each dose group. The demographic characteristics
162
of the subjects are presented in Table 2. There were no significant differences in demographic
163
characteristics such as age, height, weight, and BMI between the two randomized treatment
164
groups (all p-values > 0.05, Student’s t-test).
165 166
PK analysis
167
The PK characteristics of each second-line anti-TB drug regimen after concomitant
168
administration are presented in Table 3 and were generally consistent with the previously
169
reported values for separate administration. Cmax and AUCτ were estimated after the five-day
8
170
dosings of each second-line anti-TB drug according to the regimens presented in Table 1.
171
Relatively larger inter-individual variation was observed for cycloserine, PAS, and
172
prothionamide than for the other second-line anti-TB drugs (Table 3, Figure 1). The observed
173
Cmax of cycloserine, moxifloxacin, streptomycin, and kanamycin was consistent with the
174
available literature data (Table 4). Prothionamide and levofloxacin showed similar t1/2 to the
175
reported t1/2 of each drug. The Tmax and t1/2 of streptomycin and kanamycin were consistent
176
with previously reported data and showed rapid absorption after IM injection (Table 3, Figure
177
2).
178
Sub-group analysis showed higher mean concentrations of cycloserine and
179
prothionamide in group 2 compared with group 1 (Figures 1A and 1C). The GMR was greater
180
than 1.0 (90% CI) for the Cmax and AUCτ values in group 2 compared with group 1 for
181
cycloserine and prothionamide (Table 5). For cycloserine, the mean CL/F rates of the two
182
groups were significantly different, with (p = 0.010) or without (p = 0.026) normalization for
183
body weight. For prothionamide, the CL/F rates of the two groups were significantly different
184
(p = 0.046), but the difference was not significant after normalization for body weight (p =
185
0.100). On the other hand, for PAS, although the mean concentrations were higher in group 2
186
than in group 1 until 3 h after dosing, the 90% CI of the GMR crossed 1.0, indicating no
187
significant differences between the two groups (Table 5, Figure 1B). For PAS, there was no
188
significant difference between the two groups with respect to CL/F rate with (p = 0.873) or
189
without (p = 0.925) normalization for body weight.
190 191
Sampling time-points For TDM of second-line anti-TB drugs
192
Table 6 shows the most reliable sampling time-points for each second-line anti-TB
193
drug. These represent the time-points at which the drug concentrations were found to have
194
the strongest correlation with Cmax or AUCτ. Based on the adjusted r2 and p-values, the
9
195
concentrations measured in samples collected at 2–3 h after dosing on day 5 correlated most
196
strongly with the Cmax values of cycloserine, PAS, prothionamide, and levofloxacin than
197
samples collected at other time-points. For levofloxacin, the concentrations of levofloxacin
198
sampled 2 h and 1.5 h after dosing were most strongly correlated with Cmax and AUCτ,
199
respectively. In the case of moxifloxacin, the concentration in the sample collected 6 h after
200
dosing was most strongly correlated with both Cmax and AUCτ. For streptomycin and
201
kanamycin, the concentrations in samples collected 1 h after dosing were most strongly
202
correlated with the Cmax of both drugs.
203 204
Discussion
205
This study shows that the measured PK characteristics of the tested second-line anti-
206
TB drugs after concomitant administration were consistent those reported for separate
207
administration (4, 7, 8). Correlation analysis between the PK parameters and the observed
208
concentrations at each time-point suggested that the time-points 1, 2.5, and 6 h post-dosing
209
are required for accurate limited-sampling TDM when all seven drugs are administered
210
concomitantly. To the best of our knowledge, the present study is the first clinical study to
211
evaluate the PK characteristics of second-line anti-TB drugs in the context of the commonly
212
used regimens for MDR TB with multiple doses.
213
In general, the results of this study correspond with the PK characteristics previously
214
reported for single drug administrations (4, 7, 8). However, compared to single drug
215
administration, some differences in PK characteristics were observed when concomitantly
216
administering the drugs. For example, although the observed mean Cmax for cycloserine
217
corresponded with the literature data (9), the median Tmax was 2 h later than the known range
218
of Tmax (4). For PAS, the mean Cmax in all of the subjects was ~ 28% greater than the reported
219
mean Cmax (10), which is thought to be due to the 32% larger dose used in this study;
10
220
however, the mean AUCτ was ~ 11% smaller than the reported value of AUC0-12hr after the
221
administration of 4 g twice daily (10). In the case of prothionamide, the observed mean Cmax
222
and AUCτ in all subjects was ~ 2-fold greater than the previously reported values of each
223
parameter, although the mean dose used in the previous literature was 55.3% larger than in
224
the present study (11). For levofloxacin, the mean Cmax was more than 17.5% smaller than the
225
reference Cmax (4), though for moxifloxacin, the mean Cmax was consistent with previous data
226
(4). The mean Cmax, Tmax, and t1/2 of streptomycin and kanamycin were within the previously
227
reported ranges (4, 8).
228
Several factors may have contributed to the fact that the mean concentrations of
229
cycloserine and prothionamide, and the GMRs (90% CI) of the Cmax and AUCτ of cycloserine
230
and prothionamide were higher in group 2 than in group 1 (Table 5, Figure 1). For example,
231
the estimated CL/F (oral clearance) for cycloserine in group 1 was 2-fold larger than for
232
group 2; as the elimination of cycloserine is dependent on renal clearance (8), differences in
233
the bioavailability could be a possible explanation considering both groups consisted of
234
healthy subjects without any abnormalities in renal functions. Additionally, the apparent
235
volume of distribution (Vd/F) for cycloserine was 1.6-fold larger in group 1 than in group 2.
236
Likewise, the CL/F and the Vd/F of prothionamide in group 1 were 1.5 and 1.4-fold larger
237
than those in group 2, respectively. Therefore, the difference in the bioavailability of
238
cycloserine and prothionamide was most likely caused by the different drugs in each group.
239
In the case of PAS, the mean concentrations in group 2 were higher than in group 1 until 3 h
240
after dosing, although the 90% CI of the Cmax and AUCτ suggests no significant differences
241
between the two groups. The absorption of PAS granules, which are enteric coated and
242
designed for sustained release, might have been affected by the concomitantly administered
243
drugs (8), although further studies with a larger number of subjects are needed for precise
244
comparisons.
11
245
Drug interactions among second-line anti-TB drugs are not well-known, and this
246
study was not designed to assess these interactions; nonetheless, it is possible to speculate on
247
the possibility of drug interactions based on the reported PK mechanisms of the
248
concomitantly administered second-line anti-TB drugs. The PK characteristics of
249
aminoglycosides, such as streptomycin and kanamycin, are known to be similar, and more
250
than 90% is excreted unchanged in the urine (7, 8). Therefore, it is more likely that it is the
251
PK characteristics of levofloxacin and moxifloxacin that were responsible for the observed
252
differences between the two groups. However, the reported differences between moxifloxacin
253
and levofloxacin mainly concern the mechanism of metabolism rather than the effects on the
254
absorption process or bioavailability of the other drugs. Levofloxacin is primarily excreted
255
unchanged in the urine, with < 5% metabolized in the liver (12). In contrast, moxifloxacin
256
has multiple routes of elimination, with approximately 50% undergoing glucuronide and
257
sulfate conjugation in the liver, 25% excreted unchanged in the feces, and 20–25% excreted
258
unchanged in the urine (5, 13-15). Further studies on the factors that could have affected the
259
drug interactions focusing on the absorption process or bioavailability are necessary as
260
knowledge of the mechanism of metabolism is not sufficient to explain the differences
261
observed here.
262
Finally, the effects of weight can be considered as one of the factors affecting group
263
differences. Although no statistically significant difference was observed, the mean difference
264
in body weight between the two groups was 5.4 kg (p = 0.057) (Table 2). However, even after
265
normalizing the GMRs of the Cmax and AUCτ of cycloserine and prothionamide for weight,
266
the values remained larger in group 2 than in group 1. This demonstrates that weight
267
differences are less likely to be responsible for the differences in mean concentrations among
268
the studied drugs.
269
Although further studies are needed to clarify the factors responsible for the
12
270
difference between the two groups shown in this study, the work demonstrates that the
271
possibility of inter-patient differences in absorption/elimination PK characteristics should be
272
considered when treating MDR TB patients. In addition, such differences are one of the
273
reasons that TDM should be performed to confirm the presence of PK interactions and to
274
help prevent toxicity resulting from an interaction.
275
Another objective of this study was to determine the optimal sampling time-points
276
when using limited-sampling TDM. Among the second-line anti-TB drugs studied,
277
cycloserine, PAS, and prothionamide had relatively larger inter-individual variation than the
278
other drugs (Figure 1, Table 3). This suggests there is a greater need for TDM of cycloserine,
279
PAS, and prothionamide to improve the chance of safe successful therapy. For cycloserine,
280
PAS and prothionamide, target Cmax values of serum levels have been proposed (5, 8).
281
Additionally, serious central nervous system toxicity may be associated with elevated serum
282
concentrations of cycloserine (4). The activity of aminoglycosides such as streptomycin and
283
kanamycin relies on a high Cmax-to-MIC ratio due to concentration-dependent antibiotic and
284
post-antibiotic effects (7). In contrast, based on the known PK parameters and MICs of
285
fluoroquinolones, the expected pharmacodynamic parameters of moxifloxacin and
286
levofloxacin can be derived from Cmax/MIC and AUC/MIC (5). In this respect, the sampling
287
time-points can be suggested based on the adjusted r2 and p-values as presented in Table 6.
288
When using cycloserine, PAS, and prothionamide together, the sampling time-points should
289
be set at 2–3 h after dosing in regards to Cmax. When adding aminoglycosides such as
290
streptomycin and kanamycin, an additional sampling at 1 h after dosing can be helpful to
291
predict the efficacy. Adding a 6-h sampling point after dosing is needed for TDM when using
292
moxifloxacin. A 6-h sampling time-point is also useful in distinguishing between
293
malabsorption and late absorption of the other TB agents, and in regimens including
294
moxifloxacin an additional later time point might be considered. Therefore, when
13
295
administering all seven drugs evaluated in this study, at least three sampling time-points are
296
needed: 1 h, 2.5 h, and 6 h after dosing.
297
Extrapolating the results of this study to larger MDR TB patient populations may be
298
difficult due to the small number of healthy, relatively young male subjects who participated
299
in the study. However, in conjunction with the PK characteristics described in other studies,
300
the PK profiles of the second-line anti-TB drugs described here could be used to develop
301
basic guidelines for TDM.
302 303
Conclusion
304
This study provides valuable information concerning the PK characteristics of second-line
305
anti-TB drugs in commonly used multiple dosing regimens for MDR TB. Possible drug
306
interactions were observed when co-administering fluoroquinolones with cycloserine, PAS,
307
and prothionamide, and this may warrant further study in MDR TB patients. Based on the
308
results of this study it is also possible to recommend sampling time-points for use in reduced-
309
sampling TDM of second-line anti-TB drugs. It is envisaged that when treating patients in a
310
complex clinical setting such a procedure will enable optimal adjustments of dosing regimens
311
in order to overcome problems such as slow response to treatment, toxicity, or suspicious
312
drug interactions.
313 314
Funding
315
This work was supported by a grant from the Korean Health Technology R&D Project of the
316
Ministry of Health & Welfare, Republic of Korea (HI13C0892).
317 318
Transparency declarations
14
319
The funders had no role in the study’s design, its data collection and analysis, its decision to
320
publish or in any preparation of the manuscript. None of the authors of the study have any
321
conflict of interest to disclose.
322 323
Author contributions
324
S.-I.P., J.O., K.J, K.-S.Y., J.-Y.C. wrote the manuscript; J.Y., K.J., J.S.P., J.H.L., I.-J.J., J.-Y.C.
325
designed the research; S.-I.P., J.O., S.J.M., J.S. J.-Y.C. performed the research; and S.-I.P.,
326
J.O., J.-Y.C. analyzed the data.
327
15
328
References
329
1.
Gandhi NR, Nunn P, Dheda K, Schaaf HS, Zignol M, Van Soolingen D, Jensen P,
330
Bayona J. 2010. Multidrug-resistant and extensively drug-resistant tuberculosis: a
331
threat to global control of tuberculosis. Lancet 375:1830-1843.
332
2.
Falzon D, Jaramillo E, Schunemann HJ, Arentz M, Bauer M, Bayona J, Blanc L,
333
Caminero JA, Daley CL, Duncombe C, Fitzpatrick C, Gebhard A, Getahun H,
334
Henkens M, Holtz TH, Keravec J, Keshavjee S, Khan AJ, Kulier R, Leimane V,
335
Lienhardt C, Lu C, Mariandyshev A, Migliori GB, Mirzayev F, Mitnick CD,
336
Nunn P, Nwagboniwe G, Oxlade O, Palmero D, Pavlinac P, Quelapio MI,
337
Raviglione MC, Rich ML, Royce S, Rusch-Gerdes S, Salakaia A, Sarin R, Sculier
338
D, Varaine F, Vitoria M, Walson JL, Wares F, Weyer K, White RA, Zignol M.
339
2011. WHO guidelines for the programmatic management of drug-resistant
340
tuberculosis: 2011 update. Eur Respir J 38:516-528.
341
3.
Dartois V, Barry CE. 2010. Clinical pharmacology and lesion penetrating properties
342
of second- and third-line antituberculous agents used in the management of
343
multidrug-resistant (MDR) and extensively-drug resistant (XDR) tuberculosis. Curr
344
Clin Pharmacol 5:96-114.
345
4.
346 347
Alsultan A, Peloquin CA. 2014. Therapeutic drug monitoring in the treatment of tuberculosis: an update. Drugs 74:839-854.
5.
Garcia-Prats AJ, Donald PR, Hesseling AC, Schaaf HS. 2013. Second-line
348
antituberculosis drugs in children: a commissioned review for the World Health
349
Organization 19th Expert Committee on the Selection and Use of Essential Medicines.
350
World Health Organization, Geneva.
351
6.
Han M, Jun SH, Lee JH, Park KU, Song J, Song SH. 2013. Method for
16
352
simultaneous analysis of nine second-line anti-tuberculosis drugs using UPLC-
353
MS/MS. J Antimicrob Chemother 68:2066-2073.
354
7.
355 356
Douglas JG, McLeod M-J. 1999. Pharmacokinetic factors in the modern drug treatment of tuberculosis. Clinical pharmacokinetics 37:127-146.
8.
357
Peloquin CA. 2002. Therapeutic drug monitoring in the treatment of tuberculosis. Drugs 62:2169-2183.
358
9.
Anonymous. 2008. Cycloserine. Tuberculosis 88:100-101.
359
10.
Liwa AC, Schaaf HS, Rosenkranz B, Seifart HI, Diacon AH, Donald PR. 2013.
360
Para-aminosalicylic acid plasma concentrations in children in comparison with adults
361
after receiving a granular slow-release preparation. J Trop Pediatr 59:90-94.
362
11.
Lee HW, Kim DW, Park JH, Kim SD, Lim MS, Phapale PB, Kim EH, Park SK,
363
Yoon YR. 2009. Pharmacokinetics of prothionamide in patients with multidrug-
364
resistant tuberculosis. Int J Tuberc Lung Dis 13:1161-1166.
365
12.
Chien SC, Rogge MC, Gisclon LG, Curtin C, Wong F, Natarajan J, Williams RR,
366
Fowler CL, Cheung WK, Chow AT. 1997. Pharmacokinetic profile of levofloxacin
367
following once-daily 500-milligram oral or intravenous doses. Antimicrob Agents
368
Chemother 41:2256-2260.
369
13.
Tachibana M, Tanaka M, Masubuchi Y, Horie T. 2005. Acyl glucuronidation of
370
fluoroquinolone antibiotics by the UDP-glucuronosyltransferase 1A subfamily in
371
human liver microsomes. Drug Metab Dispos 33:803-811.
372
14.
Senggunprai L, Yoshinari K, Yamazoe Y. 2009. Selective role of sulfotransferase
373
2A1 (SULT2A1) in the N-sulfoconjugation of quinolone drugs in humans. Drug
374
Metab Dispos 37:1711-1717.
375 376
15.
Anonymous. 2008. Moxifloxacin. Tuberculosis 88:127-131.
17
377
Table and Figure Legends
378
Table 1. Dosing regimen for each study group from the day 1 morning dose to the day 5
379
morning dose.
380
Table 2. Subject demographics and characteristics
381
Table 3. Summary of pharmacokinetic parameters of second-line anti-tuberculosis drugs
382
Table 4. Summary of literature data of pharmacokinetic parameters of second-line anti-
383
tuberculosis drugs
384
Table 5. Comparison of pharmacokinetic parameters of cycloserine, PAS, and prothionamide
385
between group 1 and group 2
386
Table 6. Recommended sampling time-points as estimated by correlation analysis between
387
Cmax or AUCτ and the concentration of second-line anti-TB drugs measured in the samples at
388
each time-point on day 5
389
Figure 1. Mean plasma concentration-time profiles of (A) cycloserine, (B) p-aminosalicylic
390
acid, and (C) prothionamide after multiple administrations in group 1 (closed circles), in
391
group 2 (open circles), and in all subjects (open squares) on day 5. The error bars denote
392
standard deviations.
393
Figure 2. Mean plasma concentration-time profiles of (A) levofloxacin, (B) moxifloxacin, (C)
394
streptomycin, and (D) kanamycin after multiple administrations on day 5. The error bars
395
denote standard deviations.
1
Table 1. Dosing regimen for each study group from the day 1 morning dose to the day 5 morning dose.
Group
Regimen common to both groups
Part of regimen different between groups
1 (n = 7)
cycloserine 250 mg PO twice daily
levofloxacin 750 mg PO once daily
PAS 5.28 g PO twice daily
streptomycin 1 g IM once daily
prothionamide 250 mg PO twice daily
moxifloxacin 400 mg PO once daily
pyrazinamide 1500 mg PO once daily
kanamycin 1 g IM once daily
2 (n = 9)
IM, intramuscular; PAS, p-aminosalicylic acid; PO, per os.
2
Table 2. Subject demographics and characteristics All
Group 1
Group 2
(n = 16)
(n = 7)
(n = 9)
Age (years)
26 (21–40)
25 (22–32)
28 (21–40)
Weight (kg)
66.3 (57.9–81.0)
69.3 (61–81)
63.9 (57.9–68.7)
Height (cm)
171.3 (161.0–178.0)
173.0 (167.0–178.0)
170.0 (161.0–177.0)
BMI (kg/m2)
22.5 (19.6–25.9)
23.2 (20.6–25.9)
22.0 (19.6–24.0)
Characteristic
Data are presented as arithmetic mean (range) except for age, for which median (range) is presented.
3
Table 3. Summary of pharmacokinetic parameters of second-line anti-tuberculosis drugs cycloserine
PAS
prothionamide
levofloxacin
moxifloxacin
streptomycin
kanamycin
Tmax (h)
4.0 [2.0-6.0]
2.5 [0.5-6.0]
3.0 [1.5-6.0]
4.0 [3.0-6.0]
4.0 [1.0-6.0]
1.0 [0.5-1.5]
1.0 [0.5-2.5]
Cmax (mg/L)
24.9 ± 9.9
65.9 ± 32.2
5.3 ± 1.9
6.6 ± 1.0
4.7 ± 1.5
42.0 ± 10.8
34.5 ± 4.3
(CV %)
(39.8)
(48.9)
(35.9)
(15.2)
(31.7)
(25.8)
(12.5)
242.3 ± 99.8
326.5 ± 152.5
22.1 ± 7.9
64.4 ± 19.0
54.2 ± 14.0
196.7 ± 25.6
153.5 ± 25.7
(41.2)
(46.7)
(35.8)
(29.6)
(25.9)
(13.0)
(16.7)
20.7 ± 11.9
12.8 ± 4.7
12.3 ± 2.8 7.8 ± 1.9 (23.9)
5.2 ± 0.7 (12.9)
6.7 ± 0.9 (14.2)
(57.6)
(36.7)
(22.8)
AUCτ (mg·h/L) (CV %) CL/F (L/h)
1.2 ± 0.5 (41.6)
t1/2 (h)
20.3 ± 13.9
1.8 ± 0.5
2.4 ± 0.5
8.2 ± 3.3
13.7 ± 2.6
3.1 ± 0.4
3.2 ± 1.0
Vd/F (L)
10.7 ± 4.4
16.6 ± 13.9
41.0 ± 14.6
117.1 ± 23.9
103.8 ± 25.7
22.6 ± 2.8
30.4 ± 9.8
Data are presented as arithmetic mean ± SD except for Tmax, for which median [min–max] is presented. AUCτ, area under the plasma concentration-time curve within a dosing interval at steady state; CL/F, apparent clearance at steady state; Cmax,
4
maximum plasma concentration at steady state; CV, coefficient of variation; PAS, p-aminosalicylic acid; t1/2, terminal elimination half-life; Tmax, time to reach maximum blood concentration at steady state; Vd/F, apparent volume of distribution.
5
Table 4. Summary of literature data of pharmacokinetic parameters of second-line anti-tuberculosis drugs. cycloserine
PAS
prothionamide
levofloxacin
moxifloxacin
streptomycin
kanamycin
reference
(4) a
(9)
(10)
(11)
(4) a
(4) a
(4, 8) a
(4) a
dosage
250–500 mg QD or BIW
250 mg BID
4 g BID
250 mg or 375 mg BID (mean dose 5.9 mg/kg)
500–1,000 mg QD
400 mg QD
15 mg/kg QD
15 mg/kg QD
Tmax (h)
2
5.2
3.4
1–2
1–2
0.5–1.5
0.5–1.5
Cmax (mg/L)
20–35
51.3
2.5
8–13
3–5
35–45 b
35–45 b
368
11.3 9
7
3
3
25–30
AUCτ (mg·h/L) t1/2 (h)
7
10
3
a
‘Normal’ values of the pharmacokinetic parameters for standard doses of drugs.
b
Calculated Cmax using linear regression to 1 h post IM dose.
AUCτ, area under the plasma concentration-time curve within a dosing interval at steady state; BID, twice daily; BIW, twice weekly; Cmax, maximum plasma concentration at steady state; IM, intramuscular; PAS, p-aminosalicylic acid; QD, every day; t1/2, terminal elimination halflife; Tmax, time to reach the maximum blood concentration at steady state.
6
Table 5. Comparison of pharmacokinetic parameters of cycloserine, PAS, and prothionamide between group 1 and group 2 Cycloserine group 1
group 2
Tmax (h)
4.0 [3.0–6.0]
2.5 [2.0–4.0]
Cmax (mg/L)
18.6 ± 7.7
29.9 ± 8.8
PAS GMR
group 1
group 2
4.0 [2.0–6.0]
1.8 [0.5–4.0]
54.9 ± 28.0
75.6 ± 34.3
323.6 ±
327.6 ±
1.66
CL/F (L/h)
0.6 ± 0.3
t1/2 (h) Vd/F (L)
group 1
group 2
3.0 [2.5–4.0]
2.3 [1.5–6.0]
4.3 ± 1.8
6.3 ± 1.5 (1.16–2.04)
1.04
287.2 ± 92.5
1.45 17.8 ± 6.3
(1.18–2.19)
GMR
1.54
(0.83–2.27)
1.61 184.5 ± 81.1
GMR
1.37
(1.24–2.22)
AUCτ (mg·h/L)
Prothionamide
25.7 ± 7.6
186.0
152.7
(0.52–2.06)
(1.08–1.95)
0.3 ± 0.2
6.5 ± 4.5
6.0 ± 3.2
15.3 ± 4.8
10.1 ± 3.8
16.4 ± 5.1
23.3 ± 17.9
1.7 ± 0.2
1.8 ± 0.6
2.2 ± 0.5
2.5 ± 0.6
13.5 ± 4.9
8.6 ± 2.6
15.3 ± 10.0
17.1 ± 15.7
47.7 ± 14.9
35.2 ± 12.5
Data are presented as arithmetic mean ± standard deviations. GMR of group 2 to group 1 (90% confidence interval) AUCτ, area under the plasma concentration-time curve within a dosing interval at steady state; CL/F, apparent clearance at steady state; Cmax, maximum plasma concentration at steady state; CV, coefficient of variation; GMR, geometric mean ratio; PAS, p-aminosalicylic acid; t1/2, terminal elimination half-life; Tmax, time to reach the maximum blood concentration at steady state; Vd/F, apparent volume of distribution.
7
Table 6. Recommended sampling time-points as estimated by correlation analysis between Cmax or AUCτ and the concentration of second-line anti-tuberculosis drugs measured in the samples at each time-point on day 5
cycloserine
PAS
prothionamide
levofloxacin
moxifloxacin
streptomycin
kanamycin
a
p ≤ 0.001
b
p ≤ 0.05
Sample time for Cmax (h)
Sample time for AUCτ (h)
(adjusted r2)
(adjusted r2)
2
4
(0.939) a
(0.968) a
3
3
(0.891) a
(0.906) a
2.5
4
(0.602) a
(0.565) a
2
1.5
(0.917) a
(0.877) a
6
6
(0.523) a
(0.849) a
1
6
(0.928) a
(0.918) a
1
4
(0.858) a
(0.937) a
8
AUCτ, area under the plasma concentration-time curve from time within a dosing interval at steady state; Cmax, maximum plasma concentration of drug at steady state; PAS, p-aminosalicylic acid.