Pharmacokinetic/pharmacodynamic-based optimization of levofloxacin administration in the treatment of MDR-TB

Transcription

1 J Antimicrob Chemother 2016; 71: doi: /jac/dkw164 Advance Access publication 26 May 2016 Pharmacokinetic/pharmacodynamic-based optimization of levofloxacin administration in the treatment of MDR-TB Samiksha Ghimire 1, Natasha van t Boveneind-Vrubleuskaya 1,2, Onno W. Akkerman 3,4, Wiel C. M. de Lange 3,4, Dick van Soolingen 5,6, Jos G. W. Kosterink 1,7, Tjip S. van der Werf 4,8, Bob Wilffert 1,7, Daniel J. Touw 1 and Jan-Willem C. Alffenaar 1 * 1 University of Groningen, University Medical Center Groningen, Department of Clinical Pharmacy and Pharmacology, Groningen, The Netherlands; 2 Department of Public Health TB Control, Metropolitan Public Health Service Haaglanden, The Hague, The Netherlands; 3 University of Groningen, University Medical Center Groningen, Tuberculosis Centre Beatrixoord, Haren, The Netherlands; 4 University of Groningen, University Medical Center Groningen, Department of Pulmonary Diseases and Tuberculosis, Groningen, The Netherlands; 5 National Tuberculosis Reference Laboratory, National Institute for Public Health and the Environment (RIVM), Bilthoven, The Netherlands; 6 Radboud University Nijmegen Medical Center, Departments of Pulmonary Diseases and Medical Microbiology, Nijmegen, The Netherlands; 7 University of Groningen, Department of Pharmacy, Section of Pharmacotherapy and Pharmaceutical Care, Groningen, The Netherlands; 8 University of Groningen, University Medical Center Groningen, Department of Internal Medicine, Groningen, The Netherlands *Corresponding author. Tel: ; Fax: ; j.w.c.alffenaar@umcg.nl Authors contributed equally. The emergence of MDR-TB and XDR-TB has complicated TB treatment success. Among many factors that contribute to the development of resistance, low drug exposure is not the least important. This review summarizes the available information on pharmacokinetic properties of levofloxacin in relation to microbial susceptibilities, in order to optimize the dose and make general treatment recommendations. A total of 37 studies on adult (32 studies) and paediatric (5 studies) MDR-TB patients were included. Among the 32 adult studies, 19 were on susceptibility of Mycobacterium tuberculosis isolates to levofloxacin by MIC, 1 was on susceptibility of M. tuberculosis isolates to levofloxacin by MBC, 1 was on susceptibility of M. tuberculosis isolates to levofloxacin by mutant prevention concentration and 4 were on pharmacokinetics of levofloxacin, and 7 others were included. Likewise, out of five studies on children, two dealt with levofloxacin pharmacokinetic parameters, one reviewed CSF concentrations and two dealt with background information. In adult MDR-TB patients, standard dosing of once-daily 1000 mg levofloxacin in TB treatment did not consistently attain the target concentration (i.e. fauc/mic.100 and fauc/mbc.100) in 80% of the patients with MIC and MBC of 1 mg/l, leaving them at risk of developing drug resistance. However, with an MIC of 0.5 mg/l, 100% of the patients achieved the target concentration. Similarly, paediatric patients failed consistently in achieving given pharmacokinetic/pharmacodynamic targets due to age-related differences, demanding a shift towards once daily dosing of mg/kg. Therefore, we recommend therapeutic drug monitoring for patients with strains having MICs of 0.5 mg/l and suggest revising thecut-offvaluefrom2to1mg/l. Introduction TB is one of the world s deadliest communicable diseases. 1 3 The WHO estimated that 9.6 million people developed TB and 1.5 million people died from this disease in 2014, of whom were HIV positive. 4 The emergence and spread of MDR-TB and XDR-TB have threatened the success of TB programmes globally. 5,6 The WHO has recommended fluoroquinolones along with injectable aminoglycosides in the treatment of MDR-TB. 7 9 Yet the treatment success rates of MDR-TB vary between 36% and 79%, thereby increasing the prevalence of XDR-TB in countries burdened with MDR-TB. 6,10,11 To tackle the growing problems of drug resistance, either new drugs have to be developed or the dose of currently used therapy should be optimized, or both. 2,12 14 Recent advances in the treatment of MDR-TB include approval of two new drugs, delamanid and bedaquiline, 15,16 and other new drugs in the development pipeline shed the light of hope on the present situation. 17 Higher success rates have been observed in more individualized programmes that include therapeutic drug monitoring (TDM). 3,18 20 Among many recognized risk factors for development of drug resistance, low drug exposure in particular has been a subject of debate. 6,21 Low drug exposure could indeed explain a large proportion of therapy failure. 1,3,22 In addition, studies have revealed that in TB patients with other comorbidities, such as HIV and diabetes, drug absorption and other pharmacokinetic (PK) parameters, such as protein binding, distribution, metabolism and elimination, may be altered, resulting in less successful treatment outcomes and # The Author Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please journals.permissions@oup.com 2691

2 2-fold higher death rates In such situations, TDM can help clinicians to make informed dosing decisions. 2,3,21,28,29 Fluoroquinolones and injectable aminoglycosides are important drugs in the combination treatment regimen of MDR-TB. 30 Of the available fluoroquinolones, levofloxacin and moxifloxacin are the preferred choices due to their strong activity against Mycobacterium tuberculosis, once-daily dosing and limited adverse effects. Although moxifloxacin has the advantage of a two-stage resistance mechanism, 12,31 levofloxacin has multiple advantages in that it has less QT interval-related toxicity, has equivalent efficacy to moxifloxacin for treating MDR-TB (similar treatment success rate, sputum conversion rate, rate of adverse drug reactions). Finally, levofloxacin is more widely available, 32 especially in low- and middle-income countries. In a recent study, it was observed that development of drug resistance against fluoroquinolones occurred in 12% of patients while receiving standard dosages. 22,33 Pharmacokinetic variability could well explain the acquired drug resistance. 33 Since TDM could help physicians to attain given PK/pharmacodynamic (PD) targets for levofloxacin, we conducted a literature search on the PK/PD of levofloxacin to support TDM in order to reach a given target. Methods All available articles evaluating the use of levofloxacin in treatment of TB in humans were retrieved through PubMed using a specific search strategy, described below. Articles concerning the in vitro activity and/or efficacy of levofloxacin against M. tuberculosis and clinical studies on PK monitoring strategy and parameters of levofloxacin were reviewed. Articles that provided information on the in vitro activity and/or efficacy of levofloxacin against M. tuberculosis were divided into two groups. The first group focused on the MIC to determine the optimal dose for killing the dominant population of M. tuberculosis, whereas the second group focused on the MBC. MIC is the lowest concentration of drug that inhibits.99% of the colonies growing on drug-free control. MICs at which 50% and 90% of the isolates are inhibited are known as MIC 50 and MIC 90, respectively. MBC is defined as the lowest drug concentration that decreases the bacterial population by 2 or more log 10 units within the same period of incubation, 34 whereas mutant prevention concentration (MPC) is a measure of the susceptibility of the mutant sub-population. 35 Search terms A PubMed search (October 2014) of relevant articles was conducted to retrieve the information on TDM for levofloxacin used to treat TB in adult MDR-TB patients. Likewise, an extended PubMed search was performed (February 2016) on paediatric MDR-TB patients due to wide availability of literature published recently. The following search terms were used with filters set to full text, English: ( Mycobacterium tuberculosis [Mesh] OR Tuberculosis [Mesh] OR tuberculosis[tw]) AND ( Levofloxacin [MeSH] OR levofloxacin[tw] OR lfx[tw]); ( Mycobacterium tuberculosis [Mesh] OR Tuberculosis [MeSH] OR tuberculosis [tw]) AND ( Levofloxacin [MeSH] OR levofloxacin[tw] OR lfx[tw]) AND ( Pharmacokinetics [MeSH] OR pharmacokinetics[tw] OR concentration[tw] OR pharmacodynamics[tw] OR therapeutic effect*[tw]); Pharmacokinetics AND Pharmacodynamics AND Levofloxacin AND Tuberculosis. Likewise, for the paediatric section, the search term used was Pharmacokinetics AND Children AND Levofloxacin with filters set to full text and English. The reference lists of identified articles were manually searched for pertinent articles not identified in the electronic search. Selection criteria Identified articles were included if the activity of levofloxacin against susceptible and MDR-TB was evaluated by in vitro, in vivo and clinical studies. In vitro studies were included if methods of testing susceptibility of M. tuberculosis to levofloxacin were adequately described. This drug susceptibility testing (DST) included different methods, such as radiometric Bactec, microplate Alamar blue assay (MABA), proportion methods and broth dilution and microdilution methods in different media. Case reports and articles describing diseases caused by species of mycobacteria other than M. tuberculosis were excluded. Additionally, articles on bioanalytical procedures and assays for the measurement of plasma concentrations of levofloxacin were excluded. Methods of data collection Two reviewers independently collected data from the selected literature and compared with avoid differences. Data analysis strategy Data were extracted from included articles and were tabulated (Tables 1, 2 and 3). For comparison of results (from Tables 2 and 3), mean values were estimated from the given median values, range and size of the sample in different studies by using the formula described by Hozo et al. 36 The AUC for free drugs (unbound to plasma protein) was calculated by multiplying the individual AUCs by the percentage of protein-free drug fraction; 37,38 protein binding for levofloxacin was 40% (range 24% 40%), as published in the literature. 37 The free fractions of this variable (fauc) were divided by corresponding MIC values. To categorize MIC values as susceptible and resistant, in vitro studies on.100 strains of M. tuberculosis were compared to avoid any bias (Table 1). Results In total, 248 articles were retrieved from the initial search. After removing duplicates, 190 articles were selected for screening based on title and abstract, after which only 40 relevant full-text articles were assessed for evaluation. Ultimately, 37 articles were included, describing in vitro susceptibility testing of levofloxacin against MDR-TB strains and clinical studies on PK monitoring strategy and parameters, on both adult and paediatric MDR-TB patients (Figure 1). Out of 32 studies evaluating the use of levofloxacin in the treatment of adult MDR-TB, 19 described the susceptibility of M. tuberculosis isolates to levofloxacin by MIC, 1 described the susceptibility of M. tuberculosis isolates to levofloxacin by MBC and 1 described the susceptibility of M. tuberculosis isolates to levofloxacin by MPC. Four studies revealed the PK of levofloxacin and seven provided background information. Since only one study presented data for clearance and volume of distribution, these parameters were excluded from comparison. Among five articles on children, two presented PK of levofloxacin in plasma, one dealt with CSF of MDR-TB meningitis (MDR-TBM) patients and two provided background information. In vitro studies MIC The search strategy on articles exploring in vitro susceptibility testing is presented in Figure 1. Most studies performed DST against M. tuberculosis by determining the MIC using different available 2692

3 2693 Table 1. In vitro susceptibility testing of M. tuberculosis isolates to levofloxacin by MIC Reference Strain n Dose (mg) Method Range a (mg/l) MBC (mg/l) MIC (mg/l) MIC 90 (mg/l) 32 CI 90 (MDR-TB) 750 daily ACM 2 CC ND ND ND 51 CI 169 DR-TB (7 levofloxacin resistant) NA IPM on LJ (ILJ) and REMA 2 (ILJ); (REMA); cut-off 0.50 ND 4 to 16 (levofloxacin-resistant strains) ND 46 CI 102 (XDR and MDR-TB) NA APM susceptible at 1.0 and resistant at 2.0 ND 1 CC ND NA PM ND 0.5 ND 39 CI 21 (MDR-TB and DS-TB) 40 CI 57 (19 ofloxacin NA APM 0.5, 1.0, 4.0, 8.0; ND 0.5 ND resistant and 38 2 (CC) ofloxacin susceptible) 78 CI 68 (38 ofloxacin NA BAC ; 2 ND (ofloxacin susceptible); ND resistant and 30 (cut-off MIC (SM in gyra);.32 (DM ofloxacin resistance) in gyra) susceptible) 79 CI 62 (18 HLR, 42 LLR, 2 NA ACM 1 (LLR) and 10 (HLR) ND formation of colonies ND levofloxacin susceptible) 45 CI 162 NA standard MIC 0.5, 1, 2 ND 1 (for 131 CI) ND method 80 CI 8 NA MABA ND,1.0 ND 47 CI 420 SDAP ; cut-off susceptibility ND MIC.1.0 rate¼98.6% 0.5 to 8; 14 (for fluoroquinolone-resistant strains) 48 CI 141 (62 FS, 33 MDR, 46 ODRP) NA SDAP ; cut-off MIC.1.0 ND FS¼1.0; MDR¼1 to 4; ODRP¼ CI 101 (MDR-TB) NA M-MTT method; PM with MB 7H ; 4.0; cut-off MIC.1.0 ND MIC 50 and MIC 90 ¼0.50 and CI 243 MB 7H , 0.125, 0.25, 0.5, 1, 2, 4, 8, 16; 2.0 (for 5 resistant strains) MIC 50 and MIC 90 ¼0.25 and 0.5 cut-off MIC CI 55 NA MB 7H11 0.5, 1, 2, 4, 8, 16, 32, 64, 128 ND 0.5 (for 45 strains); 1 (for 5 strains); 2 (for 5 strains) MIC 50 and MIC 90 ¼0.5 and 2 43 CI 250 NA APM ND MIC 50 and MIC 90 ¼0.5 and 1 50 CI 135 NA BB 0.25, 0.5, 1.0, 2 susceptibility rate¼99.9% at 1; one patient MIC¼16 ND Continued Review JAC

4 Table 1. Continued Dose (mg) Method Range a (mg/l) MBC (mg/l) MIC (mg/l) MIC 90 (mg/l) Reference Strain n ND MIC50 and MIC90¼0.5 and 8 44 CI 40 (MDR-TB) NA ADM ; cut-off MIC APM ND 1 ND 82 fluoroquinoloneresistant and - susceptible MTB ND ; susceptible ECOFF 0.5 fauc0 24/MIC 45 (45 222) for 90 CI (protein binding¼40%) 37 CI 90 (24 MDR-TB) Middlebrook 7H10 medium ACM, absolute concentration method; CC, critical concentration; NA, not applicable; ND, not determined; MTB, M. tuberculosis; BAC, Bactec MGIT 960 system; SM, strains with single mutation; DM, strains with double mutations; LLR, low-level resistant; HLR, high-level resistant; SDAP, serial dilutions on agar plates; FS, fully susceptible; ODRP, other drug-resistant patterns; ADM, agar dilution method, 7H11 medium; BB, Bactec 7H12 broth; PM, proportion method; MB, Middlebrook; gyra, gene encoding DNA gyrase A; DR, drug resistant; IPM, indirect proportion method; LJ, Lowenstein Jensen; REMA, microplate colorimetricmethodusingresazurin;apm,agarproportionmethod;ds,drugsusceptible; M-MTT, microplate-based MTT. methods in different culture media. Table 1 shows the susceptibilities of drug-resistant M. tuberculosis strains to levofloxacin. Different concentrations of levofloxacin were tested in order to determine the MIC for both resistant and susceptible strains. Final concentrations of levofloxacin tested ranged between and 512 mg/l, following 2-fold dilutions. In most of the studies, susceptible strains had an MIC of 1 mg/l. 37,39 41 The MIC 50 was 0.5 mg/l and the MIC 90.1 mg/l. 37,39 41 Eight articles (n.100) that presented MIC values were compared (Table 4). 43,45 51 The inhibitory activity of levofloxacin against the strains studied is summarized in Table 4. MBC Mor et al. 34 studied the inhibitory and bactericidal activities of levofloxacin against three strains of M. tuberculosis in vitro. The MIC of levofloxacin for all three strains was 0.5 mg/l, whereas MBC was 1 mg/l for two strains and 0.50 mg/l for one strain, resulting in an MBC/MIC ratio of 2 or 1. Clinical studies PK/PD of levofloxacin in adult MDR-TB patients Mpagama et al. 52 performed an experimental study to measure the plasma concentration of levofloxacin in 25 patients with MDR-TB in Tanzania. Patients received 750 mg of levofloxacin given orally once daily in a standardized treatment regimen. After 14 days of standardized treatment with levofloxacin, drug concentrations were measured 2 h post-medication (C2). The mean levofloxacin concentration ( mg/l) was lower than the expected range ( mg/l) for 13 Tanzanian patients. The C2/MIC ratio was found to be The mean values presented in this section are calculated from the given median values in the literature by using the formula described by Hojo et al. 36 The authors concluded that low plasma concentrations of levofloxacin could be due to the single timepoint of plasma withdrawal (C2) in the dosing interval, the treatment schedule (2 weeks) and the dose itself. Once-daily dosing of 1000 mg of levofloxacin has been proved to be more effective than 750 mg in many studies. Despite these limitations, this study provides important information on plasma drug concentrations relative to the quantitative susceptibility in a standard MDR-TB treatment regimen. A summary of PK properties of levofloxacin is shown in Table 2. Two randomized controlled trials (RCTs) and one randomized open-label trial were identified that included AUC/MIC ratio as a predictor of efficacy for levofloxacin. 22,38,53,54 The body weight of the patients included in the clinical studies ranged from 45 to 66 kg. Johnson et al. 53 conducted a randomized, open-label trial to study early bactericidal activity (EBA) of levofloxacin with a daily dose of 1000 mg in 10 TB patients for 7 days. Mean fauc/ MIC and C max /MIC ratio were and 20.7, respectively. However, due to the small sample size, this study only adds to the evidence on efficacy using EBA, but not to that of long-term treatment outcome. An RCT by Peloquin et al. 54 was completed in 10 patients with TB. Patients were randomized to receive 7 days of oral isoniazid (300 mg) (standard positive comparator for EBA studies), levofloxacin (1000 mg), gatifloxacin (400 mg) or moxifloxacin (400 mg). 2694

5 Table 2. Drug concentrations in plasma Reference Type n Dose (mg) Study type Median body weight (kg) PK monitoring strategy Analytical procedure Result values (mg/l) MIC test procedure PK/PD parameter 52 MDR-TB daily experimental C2 BA C2+SD (mg/l)¼ * 53 TB daily RCT 55 ( ) C0, C1, C2, C4, C8, C12, C18, C24; protein binding¼40% 38 TBM 15 (500/12 h) twice daily as part of standard treatment RCT median¼48 (min, max¼31, 60) lumbar puncture; 0 8, 8 16 and h after drug administration; CSF was paired with blood specimen 54 PTB daily RCT 56 (41 66) C0, C1, C2, C4, C8, C12, C18, C24; protein binding¼25% HPLC AUC 0 24 (mg.h/l)¼129.1 ( )** and C max (mg/l)¼15.6 ( )** HPLC/MS and HPLC/MS/ MS AUC 0 24 (mg.h/l) in plasma¼ 155 (81.8, 284)* and AUC 0 24 (mg.h/l) in CSF¼94.1 (53.1, 208)* ofloxacin MIC determined, correlated to levofloxacin by potency relationship HPLC AUC 0 24 (mg.h/l)¼129 ( )** and C max (mg/l)¼15.55 ( )** 0.75 (0.25 1);** ECOFF 0.5 MSP C2/MIC¼ * MIC 90 ¼1 APM and BRM C max /MIC¼15.6 ( )** and AUC 0 24 / MIC¼129.1 ( )** actual MIC¼0.5** and MIC 90 ¼1** 1% proportion method on Lowenstein Jensen medium BRM CSF AUC 0 24 /plasma AUC 0 24 ¼0.74 AUC 0 24 /actual MIC¼258 and AUC 0 24 / MIC 90 ¼129 C2, plasma drug concentration at 2 h after medication administration; ROT, randomized open label trial; BA, bioassay; BRM, Bactec radiometric method; MSP, MYCOTB Sensititre plates; PTB, pulmonary TB; *, mean values; **, median values. References 53 and 54 are about the same study. JAC 2695

6 2696 Table 3. Drug concentrations in plasma of paediatric MDR-TB patients Reference Type n Dose Study type Age (years) PK monitoring strategy Analytical procedure Result values (mg/l) MIC test procedure PK/PD 55 MDR-TB prophylaxis or treatment mg/kg od PCIPSS,8 C0, C1, C2, C4, C6, C8 HPLC/ MS/MS AUC 0 1 (mg.h/l) median (IQR)¼32.92 ( ) and C max (mg/l) median (IQR)¼6.79 ( ) 1 and 0.5 published MIC 90 estimates when MIC is 1 mg/l: mean (SD) AUC 0 24 / MIC¼43.8 (13.3) and mean (SD) C max /MIC¼6.5 (2.0) when MIC is 0.5 mg/l: AUC 0 24 /MIC¼87.6 and mean (SD) C max /MIC¼13.1 (4) 58 FSM: MDR-TB or presumed LTBI RMI: presumed to be infectious MDR-TB 33 (8 MDR- TB, 25 presumed LTBI) 10 mg/kg (age.5 years); mg/kg (age 5 years) mg/kg (age 5 years); mg/kg (age,5 years) C1, C2, C6 HPLC mean C max (mg/l)¼6.09 (for age.5 years) and 8.0 (for age 5 years) and mean+sd AUC 0 6 (mg.h/l)¼ C0, C1, C2, C6 HPLC mean C max (mg/l)¼8.13 (for age.5 years) and 8.0 (for age 5 years) and mean+sd AUC 0 6 (mg.h/l)¼ , 0.5,1 and 2 refer to the Results section refer to the Results section PCIPSS, prospective cross-over intensive PK sampling study; C max, plasma drug concentration 2 h after medication administration; od, once daily.

7 JAC 248 articles retrieved from PubMed search with filters on to full text, English and date customized until 31/10/2014 for other sections and 29/02/2016 for paediatric MDR-TB patients using the search terms mentioned above 190 articles selected for screening of title and abstract 58 duplicates removed Articles excluded articles selected - Articles describing diseases caused by other bacteria and other species of genus Mycobacterium except Mycobacterium tuberculosis - Articles describing anti-tb drugs other than levofloxacin - Articles on animals - Articles on different methods and assays for measuring plasma concentration of levofloxacin 3 articles excluded: did not provide relevant information relating to levofloxacin Reading complete papers 3 discarded 37 selected Figure 1. Flow chart of the selection procedure. Table 4. Comparison of levofloxacin MIC values from eight studies with n.100 Reference n MIC (mg/l) ### ### # 33 ## # 6 Total % # MIC 90 ; ## MIC 90 (1 to.4); ###,1. Compared with gatifloxacin and moxifloxacin, levofloxacin exhibited the highest maximum plasma concentrations (median mg/l; gatifloxacin 4.75 mg/l; moxifloxacin 6.13 mg/l), the largest volume of distribution (median 81 L; gatifloxacin 79 L; moxifloxacin 63 L) and the longest elimination half-life (median 7.4 h; gatifloxacin 5.0 h; moxifloxacin 6.5 h). The mean fauc/mic ratio (protein binding 40% for levofloxacin, using the published MIC value of 1 mg/l) was and C max /MIC 90 was ,54 In addition, from the available PK data on 10 patients, we calculated the fauc 0 24 /MIC ratio for each patient using MIC values of 0.5 and 1mg/L(Figure2) and found that 8/10 patients had low serum concentrations and an fauc 0 24 /MIC ratio of,100 at a levofloxacin dose of 1000 mg once daily (MIC 1 mg/l). A second RCT by Thwaites et al. 38 studied the PK of levofloxacin and the exposure response relationship for the efficacy of levofloxacin in patients with TBM. Fifteen patients received levofloxacin (500 mg/12 h) for the first 60 days of therapy in combination with standard TB treatment. The efficacy in the levofloxacin 2697

8 Target attainment (%) MIC (mg/l) Figure 2. Relationship between MIC and fauc and proportion of patients with response to treatment. Open circles, fauc/mic.100; filled circles, fauc/mic.50. treatment group was compared with that in the ciprofloxacin treatment group (n ¼ 15) and the gatifloxacin treatment group (n¼15). CSF penetration, calculated as the ratio of plasma AUC 0 24 to CSF AUC 0 24, was greater for levofloxacin (median 0.74; range ) than for gatifloxacin (median 0.48; range ) and ciprofloxacin (median 0.26; range ) at the doses explored. Although levofloxacin had a better CSF penetration, the fauc 0 24 /MIC of plasma was low, i.e. 93 (protein binding, 40%), with the 500 mg twice daily dose, assuming an MIC of 1 mg/l an assumption based on MIC 90. PK/PD of levofloxacin in paediatric MDR-TB patients A prospective cross-over intensive PK sampling study investigated thepkoflevofloxacin(15mg/kg)in22children(3monthsto 8 years) on either WHO s MDR-TB treatment regimen or prophylaxis therapy for 6 months. The authors documented lower AUCs and C max values in children, with mean AUC 0 1 and C max at mg.h/l and 6.58 mg/l, respectively 55 (Table 3). The estimatedmeanauc 0 24 /MIC and C max /MIC was 43.8 and 6.5, respectively, when the published MIC 90 of 1 mg/l was used. Even with a lower MIC of 0.5 mg/l, children failed to meet the clinical outcomes relating to the ratios of AUC 0 24 /MIC.125 (mean AUC 0 24 /MIC at 87.6 in this case). However, estimated C max /MIC (13.1) was well above the proposed ratio of In addition, drug clearance was more rapid in children, with a half-life of 3 4 h and an elimination rate constant (k el ) of 0.22 h 21, compared with half-lives of 4 8 h with a k el of h 21 in adults, which could be attributed to the age-related differences in elimination of drug in the children. 54,56 The authors commented that the lower levofloxacin exposure in this study compared with existing paediatric studies could stem from the differences in drug formulations and methods of administration between the studies. A review article on CSF penetration of anti-tb agents found favourable CSF concentrations of levofloxacin at relatively low doses of mg, in comparison with MIC and plasma concentrations, in MDR-TBM children. 57 This observation is in line with the study on adult MDR-TBM patients. Despite the good penetration profile, levofloxacin 20 mg/kg is recommended to ensure maximum protection (which correlates with excellent EBA), preferably with a once daily dosing scheme. 55,57 However, in HIV-co-infected MDR-TBM patients, the use of levofloxacin and other fluoroquinolones was not that encouraging, which, the author explained, could result from the lack of adequate protection provided by companion drugs. 57 Mase et al. 58 studied 50 children receiving a once-daily levofloxacin-based regimen for MDR-TB or latent TB infection (LTBI), presumed to have MDR-TB, at doses of 10 mg/kg (age.5 years) and mg/kg (age 5 years) at two different treatment centres. The clinical characteristics and levofloxacin PK parameters were studied to determine the optimal dosages and inform future dosing strategies. A levofloxacin dosage of mg/kg achieved the desired C max of.8 mg/l, whereas AUC 0 6 (mean+sd) was mg.h/l for children at the Federated States of Micronesia (FSM) centre and mg.h/l for those at the Republic of the Marshall Islands (RMI) centre. Based on PK modelling, fauc ss, 0 24 /MIC was calculated at a steady state using simulated paediatric exposures and typical MICs (0.25, 0.50, 1 and 2 mg/l) with a protein binding percentage of 25 for levofloxacin. For MIC,0.5 mg/l, high target attainment was achieved with 15 mg/kg daily dosing, whereas with MIC 0.5 mg/l, 20 mg/kg daily dosing would meet the target. The authors recommended revision of the current dosage for levofloxacin to achieve a target C max 8mg/Landadvised the use of levofloxacin at mg/kg once daily in children 2 years of age. 58 Discussion Adult MDR-TB patients The most important finding of this review is that the currently recommended dosage of levofloxacin in MDR-TB treatment is not sufficient to reach the target concentration in the majority of patients that have strains of M. tuberculosis with MIC values of.0.5 mg/l. This raises a serious concern about the development of drug resistance, treatment failure and further extension to XDR-TB. Since MDR-TB is currently treated with a combination therapy in which fluoroquinolones, including levofloxacin, are the central drugs in the treatment regimen, optimizing the dose of individual drugs (levofloxacin in this case) would allow optimal efficacy of combination therapy regimens. The dose of levofloxacin could be optimized by taking PK/PD calculations into consideration. The best activity of levofloxacin against M. tuberculosis and decreased likelihood of drug resistance are well recognized at fauc/mic ratios of.100 and C max /MIC ratios of Without exception, patients with fauc/mic ratios,100 are at a greater risk of drug resistance. 38 Therefore, this paper outlines more rational methods for designing dosages and dosing schedules of levofloxacin in MDR-TB treatment. First of all, levofloxacin doses of 1000 mg daily 53,54 and 500 mg twice daily 38 were evaluated for their ability to achieve the desired fauc/mic of 100. Levofloxacin 500 mg twice daily had an fauc/mic ratio of,100 when an MIC value of 1 mg/l was used. This clearly reveals the lack of drug exposure due to the dose-related differences in the mean steady-state fauc 0 24 and C max values between two different dosing schemes. As a result, patients taking a regimen of,500 mg twice daily are at 2698

9 JAC risk of developing acquired drug resistance. 38,52 Because fluoroquinolones have a concentration-dependent killing rate, it is expected that administration of a once-daily dose would result in better efficacy compared with administering the same total daily dosage in divided portions. 59 In addition, the study by Peloquin et al. 54 showed that 8/10 (80%) patients receiving 1000 mg of levofloxacin had low serum concentrations and an fauc 0 24 /MIC ratio of,100 (Figure 2) when an MIC value of 1 mg/l was assumed. In contrast, when an actual MIC of 0.5 mg/l was adopted, 100% of the patients had fauc/mic.100. The MIC values exhibited by strains of M. tuberculosis from each individual are termed actual MICs, whereas 1 mg/l is the established MIC 90 value. 54 It is noteworthy that an actual MIC difference of 0.5 versus 1.0 mg/l can be within the range of inter-reader variability in interpreting an MIC manual readout, or within the range of variability (1 dilution in either direction) if the assay is repeated on the same M. tuberculosis isolate. This emphasizes just how near the majority of isolates are to an MIC that is at the breakpoint of clinical efficacy. Sirgel et al. 60 compared the quinolone resistance-determining regions of gyra genes with MICs of ofloxacin and moxifloxacin for M. tuberculosis and demonstrated that the MICs of ofloxacin and levofloxacin were not equally affected by the mutations and were still within or marginally outside the normal WT distribution. Thus, conventional qualitative susceptibility testing based on a single critical concentration is not suitable for distinguishing among borderline (lowlevel), moderate-level and high-level resistance. A considerable proportion of isolates with so-called borderline susceptibility to ofloxacin in combination with inadequate drug concentration (due to individual PK variability) resulted in ineffectiveness of the drug. 61 Additionally, patients with a borderline resistant strain might benefit from dose adjustment or in-class change. 61 Therefore, MIC determination reflects the bacterial population, giving the possibility of guiding the design of treatment modalities. A well-defined clinical breakpoint should be configured to detect resistance in each patient and to individualize therapy. Despite the fact that MIC testing compares favourably with standard phenotypic DST, detection of MICs is still infrequently performed in TB endemic settings for several reasons. 62 Recently, the MYCOTB MIC plate has been made available for testing MICs for M. tuberculosis. This test, configured for determination of MICs, exhibits fair agreement with DST results using the Middlebrook 7H10 agar proportion method and Bactec MGIT 960, and we recommend its implementation in MDR treatment centres and further prevention of XDR-TB. 61 On the other hand, 80% of the patients had an fauc/mbc,100 with an MBC of 1 mg/l. Thus, with increasing MIC (.0.5 mg/l), and corresponding MBC, only around 20% of patients attained the target plasma concentration for free levofloxacin, leaving the remaining 80% on the verge of developing acquired drug resistance. This provides an important reason to worry about inter-patient variability in the PK of levofloxacin, which leads to suboptimal drug concentrations with the standardized dose in a proportion of patients and stresses the importance of TDM in improving treatment success. Based on these observations, we advocate that TDM should be performed in patients with strains of M. tuberculosis having an MIC value.0.5 mg/l to determine the optimal dose for killing bacteria and decrease the proportion of patients at risk of treatment failure due to the lack of target attainment. 63 Second, the body weight of patients ranged from 45 to 66 kg. What could be learned from this observation is that body weight could influence the achievement of the target concentration and, for a patient who weighs.70 kg, a 1000 mg dose might still not be sufficient. This aspect is supported by Alsultan et al. 64 in a recent study that showed that the use of levofloxacin dosages in the range of mg/kg gave good target attainment for MICs from 0.25 to As individualized treatment of MDR-TB depends upon reliable and valid DST, 65 we reviewed the results on 1615 strains that were tested for susceptibility to levofloxacin. The MIC of levofloxacin ranged between 0.06 and.16 mg/l (Table 4). More than half of the strains tested, i.e %, exhibited an MIC value of 1 mg/ L, whereas 20.06% were susceptible at 0.5 mg/l and 15.17% at 0.25 mg/l. This shows that the majority of isolates are susceptible (MIC 1 mg/l). Based on this, we conclude that levofloxacin is a good drug to start with for a desired response (microbial kill). However, 6.56% of the strains had an MIC of 2 mg/l. For susceptible strains a standardized dose is preferred and for resistant strains other drugs are prescribed. If strains have an MIC between 1 and 2 mg/l, i.e. 1.0, 1.25, 1.50, 1.75 and 2.0 mg/l, we recommend increasing the dose, thereby increasing the AUC or C max. 66 With increasing MIC, microbial killing is eventually lost and levofloxacin is no longer effective. 66 Therefore, for patients that have MIC values between 1 and 2 it is important to consider susceptibility breakpoints different from those in use for the selection of doses that suppress the emergence of drug resistance. This recommendation of increasing the levofloxacin dose could raise concerns about dose-related toxicities. Fortunately, the rate of adverse drug reactions of levofloxacin is still one of the lowest among fluoroquinolones. 67 There is insufficient evidence to be clear about whether adverse effects of fluoroquinolones levofloxacin in particular in the treatment of TB are highly dose dependent. The high dosage regimens of fluoroquinolone against severe bacterial infections resulted in greater rates of clinical cure and low frequency of side effects. 68,69 In addition, increasing the dose of moxifloxacin in patients with TBM appeared to be safe. 70,71 Levofloxacin dosed at 1000 mg once daily has not only shown the best EBA, but has also proved to be clinically tolerated and safe. 53,54 Nevertheless, worries about development of dose-dependent toxicity are understandable. Paediatric MDR-TB patients The results of studies on paediatric MDR-TB patients demonstrate how age-related differences in the clearance of levofloxacin could considerably affect drug exposure, thereby affecting clinical outcomes,andhowdoseextrapolationinchildrenbasedonthe established PK of levofloxacin in adults could fail to approximate drug exposures. 55,56 Levofloxacin elimination, as characterized by half-life and clearance, is age dependent. This clearly stresses the need to optimize levofloxacin treatment, by increasing either the dose or the dosing frequency in order to attain given PK/PD targets. Although fluoroquinolones have been used with caution in children due to safety concerns, the available data from different studies have not demonstrated any serious arthropathy or other severe toxicity in children. 55,72 Further studies (RCTs) are needed to evaluate the safety and PK of levofloxacin, especially in HIV-infected children and children,2 years of age, in a highdose, long-term MDR-TB regimen. 56 Despite the dearth of adequate controlled studies, failure to achieve given PD targets 2699

10 MDR-TB confirmation DST for second-line drugs* MDR-TB treatment TDM after 2 weeks (e.g. using DBS) Actual MIC not available Actual MIC available and TDM Increase dose until fauc/mic >100 -use MIC population cut-off value No dose adjustments if -fauc/mic >100 -MIC <0.5 mg/l Dose adjustments needed if -fauc/mic <100 -MIC >0.5 mg/l Figure 3. Flow chart showing the incorporation of TDM for the successful treatment of MDR-TB. *Susceptibility breakpoint results (1 mg/l for levofloxacin proposed). with current levofloxacin dosages in children creates the need to modify current and future dosing recommendations towards the top of the accepted range for levofloxacin, i.e. around 20 mg/kg, with careful monitoring of efficacy and toxicity both in MDR-TB and MDR-TBM paediatric patients. Above all, it is crucial to mention that TDM testing could be a challenge in resource-constrained settings where there is a high TB burden. Regarding the fauc/mic ratio, for example, a series of blood samples (minimum of six or seven samples) need to be collected in order to calculate fauc. However, obtaining a full concentration-time profile might not be feasible in rural clinics. 73 Therefore, limited sampling strategies could be applied to estimate the total drug exposure. 64 In addition, dried blood spot (DBS) sampling could be introduced for the measurement of drug concentrations in settings where the distance between peripheral, intermediate and central laboratories is large. DBS overcomes the problems associated with venous blood sampling, such as storage, refrigeration and transportation. 74 Samples could be collected in the second week of treatment to perform TDM. Figure 3 describes TDM incorporation for the successful treatment of MDR-TB. The limitation of this review is that articles related to the aim were detected only by searching PubMed. We did not use other search engines, such as Medline, Embase or the Cochrane database. Conclusions We provide a comprehensive summary of data on levofloxacin concentrations achieved in the plasma of TB patients and MIC values of levofloxacin for susceptible and resistant M. tuberculosis isolates. A PK/PD approach to determining the dose by considering the fauc/mic ratio for levofloxacin could help many clinicians to select the right dose for successful treatment of MDR-TB and hence prevent the emergence of XDR-TB. Based on this review, we recommend firstly TDM for patients with isolates having an MIC of 0.5 mg/l to prevent acquired drug resistance to levofloxacin, and secondly the use of standardized dosing at 1000 mg once daily rather than in divided portions. For patients with body weight 70 kg and MIC from 0.25 to 0.50 mg/l, levofloxacin dosages in the range of mg/kg are suggested. Although, theoretically, MPC is considered more sensitive than MIC in terms of preventing the emergence of resistant mutants during treatment, the MPC concept is relatively new in the field of PK/PD. Further work (e.g. a prospective study) is needed to consolidate its usefulness in MDR-TB treatment. Similarly, in paediatric patients, due to higher elimination of levofloxacin, especially for those,5 years of age, current standard dosing clearly falls short in meeting the given PK/PD targets. To reach levofloxacin exposure associated with better clinical outcomes, we recommend personalized therapeutics for levofloxacin, with mg/kg once daily dosing for children on MDR-TB and MDR-TBM treatment. 2700

11 JAC Based on these observations, we advise revising the cut-off value for levofloxacin to around 1 mg/l instead of the currently used 2 mg/l. Despite the fact that DST is still regarded as the gold standard, interpretation of results may be complicated by various factors, such as the characteristics of the microorganism, not to mention the requirements for quality laboratory techniques. Values of critical concentrations, which differentiate between WT and non-wt strains, are best used as an epidemiological tool to detect changes in drug resistance rates. 37,75 WHO guidelines for the programmatic management of drug-resistant TB, revised in 2014, decreased the critical concentration of levofloxacinto1mg/linmiddlebrook7h10and1.5mg/linmgit 960, differing from ofloxacin (2.0 mg/l in both media). 76 This revision confirms how rarely DST is performed over a quantitative range. Although there may be some expected variability among the MIC methods and media used, quantitative DST is far more useful within the context of individual PK variability than a simple qualitative S or R readout. 61,76,77 Funding This study was carried as a part of our routine work. Transparency declarations None to declare. References 1 Peloquin CA. Therapeutic drug monitoring in the treatment of tuberculosis. Drugs 2002; 62: Wilby KJ, Ensom MH, Marra F. Review of evidence for measuring drug concentrations of first-line antitubercular agents in adults. Clin Pharmacokinet 2014; 53: Alsultan A, Peloquin CA. Therapeutic drug monitoring in the treatment of tuberculosis: an update. Drugs 2014; 74: WHO. Global Tuberculosis Report bitstream/10665/191102/1/ _eng.pdf?ua=1. 5 Gandhi NR, Nunn P, Dheda K et al. Multidrug-resistant and extensively drug-resistant tuberculosis: a threat to global control of tuberculosis. Lancet 2010; 375: Lange C, Abubakar I, Alffenaar JW et al. Management of patients with multidrug-resistant/extensively drug-resistant tuberculosis in Europe: a TBNET consensus statement. Eur Respir J 2014; 44: Blumberg HM, Burman WJ, Chaisson RE et al. American Thoracic Society/ Centers for Disease Control and Prevention/Infectious Diseases Society of America: treatment of tuberculosis. Am J Respir Crit Care Med 2003; 167: Pranger AD, Alffenaar JWC, Aarnoutse RE. Fluoroquinolones, the cornerstone of treatment of drug-resistant tuberculosis: a pharmacokinetic and pharmacodynamic approach. Curr Pharm Des 2011; 17: Ziganshina LE, Squire SB. Fluoroquinolones for treating tuberculosis. Cochrane Database Syst Rev 2008; issue 1: CD Falzon D, Gandhi N, Migliori GB et al. Resistance to fluoroquinolones and second-line injectable drugs: impact on multidrug-resistant TB outcomes. Eur Respir J 2013; 42: OrensteinEW,BasuS,ShahNSet al. Treatment outcomes among patients with multidrug-resistant tuberculosis: systematic review and meta-analysis. Lancet Infect Dis 2009; 9: Gumbo T, Louie A, Deziel MR et al. Selection of a moxifloxacin dose that suppresses drug resistance in Mycobacterium tuberculosis, by use of an in vitro pharmacodynamic infection model and mathematical modeling. J Infect Dis 2004; 190: van der Paardt A, Wilffert B, Akkerman OW et al. Evaluation of macrolides for possible use against multidrug-resistant Mycobacterium tuberculosis. Eur Respir J 2015; 46: Alsaad N, Wilffert B, van Altena R et al. Potential antimicrobial agents for the treatment of multidrug-resistant tuberculosis. Eur Respir J 2014; 43: Zumla AI, Gillespie SH, Hoelscher M et al. New antituberculosis drugs, regimens, and adjunct therapies: needs, advances, and future prospects. Lancet Infect Dis 2014; 14: Skripconoka V, Danilovits M, Pehme L et al. Delamanid improves outcomes and reduces mortality in multidrug-resistant tuberculosis. Eur Respir J 2013; 41: Tiberi S, De Lorenzo S, Centis R et al. Bedaquiline in MDR/XDR-TB cases: first experience on compassionate use. Eur Respir J 2014; 43: Van Altena R, de Vries G, Haar C et al. Highly successful treatment outcome of multidrug-resistant tuberculosis in the Netherlands, Lancet Infect Dis 2015; 19: Srivastava S, Peloquin CA, Sotgiu G et al. Therapeutic drug management: is it the future of multidrug-resistant tuberculosis treatment? Eur Respir J 2013; 42: Laniado-Laborin R, Moser K, Estrada-Guzman J et al. Successful treatment of multidrug-resistant tuberculosis through a binational consortium. Eur Respir J 2011; 38 Suppl 55: Aarnoutse RE, Sturkenboom MG, Robijns K et al. An interlaboratory quality control programme for the measurement of tuberculosis drugs. Eur Respir J 2015; 46: Cegielski JP, Dalton T, Yagui M et al. Extensive drug resistance acquired during treatment of multidrug-resistant tuberculosis. Clin Infect Dis 2014; 59: Peloquin C. Use of therapeutic drug monitoring in tuberculosis patients. Chest J 2004; 126: Peloquin CA. Tuberculosis drug serum levels. Clin Infect Dis 2001; 33: Kikvidze M, Ikiashvili L. Comorbidities and MDR-TB treatment outcomes in Georgia cohort. Eur Respir J 2014; 44 Suppl 58: P Daskapan A, de Lange WC, Akkerman OW et al. The role of therapeutic drug monitoring in individualised drug dosage and exposure measurement in tuberculosis and HIV co-infection. Eur Respir J 2015; 45: Akkerman OW, van Altena R, Klinkenberg T et al. Drug concentration in lung tissue in multidrug-resistant tuberculosis. EurRespirJ2013; 42: Pasipanodya JG, McIlleron H, Burger A et al. Serum drug concentrations predictive of pulmonary tuberculosis outcomes. JInfectDis2013; 208: Pranger AD, van Altena R, Aarnoutse RE et al. Evaluation of moxifloxacin for the treatment of tuberculosis: 3 years of experience. Eur Respir J 2011; 38: Kang BH, Shim TS, Lee S et al. Current status of fluoroquinolone use for treatment of tuberculosis in Korea. Eur Respir J 2013; 42 Suppl 57: P Malik M, Drlica K. Moxifloxacin lethality against Mycobacterium tuberculosis in the presence and absence of chloramphenicol. Antimicrob Agents Chemother 2006; 50: Koh W, Lee SH, Kang YA et al. Comparison of levofloxacin versus moxifloxacin for multidrug-resistant tuberculosis. Am J Resp Crit Care 2013; 188:

12 33 Alffenaar JC, Gumbo T, Aarnoutse RE. Acquired drug resistance: we can do more than we think! Clin Infect Dis 2015; 60: Mor N, Vanderkolk J, Heifets L. Inhibitory and bactericidal activities of levofloxacin against Mycobacterium tuberculosis in vitro and in human macrophages. Antimicrob Agents Chemother 1994; 38: Rodriguez JC, Cebrian L, Lopez M et al. Mutant prevention concentration: comparison of fluoroquinolones and linezolid with Mycobacterium tuberculosis. J Antimicrob Chemother 2004; 53: Hozo SP, Djulbegovic B, Hozo I. Estimating the mean and variance from the median, range, and the size of a sample. BMC Med Res Methodol 2005; 5: Angeby KA, Jureen P, Giske CG et al. Wild-type MIC distributions of four fluoroquinolones active against Mycobacterium tuberculosis in relation to current critical concentrations and available pharmacokinetic and pharmacodynamic data. J Antimicrob Chemother 2010; 65: Thwaites GE, Bhavnani SM, Chau TTet al. Randomized pharmacokinetic and pharmacodynamic comparison of fluoroquinolones for tuberculous meningitis. Antimicrob Agents Chemother 2011; 55: Rey-Jurado E, Tudo G, de la Bellacasa JP et al. In vitro effect of threedrug combinations of antituberculous agents against multidrug-resistant Mycobacterium tuberculosis isolates. Int J Antimicrob Agents 2013; 41: Blackman A, May S, Devasia R et al. Microcolonies in fluoroquinolone agar proportion susceptibility testing of Mycobacterium tuberculosis: an indicator of drug resistance. Eur J Clin Microbiol Infect Dis 2012; 31: Rodriguez JC, Ruiz M, Climent A et al. In vitro activity of four fluoroquinolones against Mycobacterium tuberculosis. Int J Antimicrob Agents 2001; 17: Morcillo N, Di Giulio B, Testani B et al. A microplate indicator-based method for determining the susceptibility of multidrug-resistant Mycobacterium tuberculosis to antimicrobial agents. Int J Tuberc Lung Dis 2004; 8: Ruiz-Serrano MJ, Alcala L, Martinez L et al. In vitro activities of six fluoroquinolones against 250 clinical isolates of Mycobacterium tuberculosis susceptible or resistant to first-line antituberculosis drugs. Antimicrob Agents Chemother 2000; 44: Lai CC, Tan CK, Huang YT et al. Extensively drug-resistant Mycobacterium tuberculosis during a trend of decreasing drug resistance from 2000 through 2006 at a medical center in Taiwan. Clin Infect Dis 2008; 47: e Singh M, Chauhan DS, Gupta P et al. In vitro effect of fluoroquinolones against Mycobacterium tuberculosis isolates from Agra & Kanpur region of north India. Indian J Med Res 2009; 129: Ahmed I, Jabeen K, Inayat R et al. Susceptibility testing of extensively drug-resistant and pre-extensively drug-resistant Mycobacterium tuberculosis against levofloxacin, linezolid, and amoxicillin-clavulanate. Antimicrob Agents Chemother 2013; 57: Wang JY, Lee LN, Lai HC et al. Fluoroquinolone resistance in Mycobacterium tuberculosis isolates: associated genetic mutations and relationship to antimicrobial exposure. J Antimicrob Chemother 2007; 59: Huang TS, Kunin CM, Shin-Jung Lee S et al. Trends in fluoroquinolone resistance of Mycobacterium tuberculosis complex in a Taiwanese medical centre: J Antimicrob Chemother 2005; 56: Rodriguez JC, Ruiz M, Lopez M et al. In vitro activity of moxifloxacin, levofloxacin, gatifloxacin and linezolid against Mycobacterium tuberculosis. Int J Antimicrob Agents 2002; 20: Perlman DC, El Sadr WM, Heifets LB et al. Susceptibility to levofloxacin of Mycobacterium tuberculosis isolates from patients with HIV-related tuberculosis and characterization of a strain with levofloxacin monoresistance. AIDS 1997; 11: Imperiale B, Zumárraga M, Di Giulio A et al. Molecular and phenotypic characterisation of Mycobacterium tuberculosis resistant to antituberculosis drugs. Int J Tuberc Lung Dis 2013; 17: Mpagama SG, Ndusilo N, Stroup S et al. Plasma drug activity in patients on treatment for multidrug-resistant tuberculosis. Antimicrob Agents Chemother 2014; 58: Johnson J, Hadad D, Boom W et al. Early and extended early bactericidal activity of levofloxacin, gatifloxacin and moxifloxacin in pulmonary tuberculosis. Int J Tuberc Lung Dis 2006; 10: Peloquin CA, Hadad DJ, Molino LP et al. Population pharmacokinetics of levofloxacin, gatifloxacin, and moxifloxacin in adults with pulmonary tuberculosis. Antimicrob Agents Chemother 2008; 52: Thee S, Garcia-Prats AJ, McIlleron HM et al. Pharmacokinetics of ofloxacin and levofloxacin for prevention and treatment of multidrugresistant tuberculosis in children. Antimicrob Agents Chemother 2014; 58: Thee S, Garcia-Prats A, Donald P et al. Fluoroquinolones for the treatment of tuberculosis in children. Tuberculosis 2015; 95: Donald P. Cerebrospinal fluid concentrations of antituberculosis agents in adults and children. Tuberculosis 2010; 90: Mase SR, Jereb JA, Gonzalez D et al. Pharmacokinetics and dosing of levofloxacin in children treated for active or latent multidrug-resistant tuberculosis, Federated States of Micronesia and Republic of the Marshall Islands. Pediatr Infect Dis J 2016; 35: Chien SC, Wong FA, Fowler CL et al. Double-blind evaluation of the safety and pharmacokinetics of multiple oral once-daily 750-milligram and 1-gram doses of levofloxacin in healthy volunteers. Antimicrob Agents Chemother 1998; 42: Sirgel FA, Warren RM, Streicher EM et al. gyra mutations and phenotypic susceptibility levels to ofloxacin and moxifloxacin in clinical isolates of Mycobacterium tuberculosis. J Antimicrob Chemother 2012; 67: Heysell SK, Ahmed S, Ferdous SS et al. Quantitative drug-susceptibility in patients treated for multidrug-resistant tuberculosis in Bangladesh: implications for regimen choice. PLoS One 2015; 10: e Turnidge J. Pharmacokinetics and pharmacodynamics of fluoroquinolones. Drugs 1999; 58: Gumbo T. New susceptibility breakpoints for first-line antituberculosis drugs based on antimicrobial pharmacokinetic/pharmacodynamic science and population pharmacokinetic variability. Antimicrob Agents Chemother 2010; 54: Alsultan A, An G, Peloquin CA. Limited sampling strategy and target attainment analysis for levofloxacin in patients with tuberculosis. Antimicrob Agents Chemother 2015; 59: Bastos ML, Hussain H, Weyer K et al. Treatment outcomes of patients with multidrug-resistant and extensively drug-resistant tuberculosis according to drug susceptibility testing to first- and second-line drugs: an individual patient data meta-analysis. Clin Infect Dis. 2014; 59: Gumbo T, Pasipanodya JG, Wash P et al. Redefining multidrug-resistant tuberculosis based on clinical response to combination therapy. Antimicrob Agents Chemother 2014; 58: Carbon C. Comparison of side effects of levofloxacin versus other fluoroquinolones. Chemotherapy 2001; 47 Suppl 3: 9 14; discussion Modai J. High-dose intravenous fluoroquinolones in the treatment of severe infections. J Chemother 1999; 11: Senneville E, Poissy J, Legout L et al. Safety of prolonged high-dose levofloxacin therapy for bone infections. J Chemother 2007; 19: