American Journal of Respiratory and Critical Care Medicine

Tuberculosis continues to be a major cause of morbidity and mortality in the world. The expansion of tuberculosis control programs has been limited by the lengthy and cumbersome nature of current chemotherapeutic regimens. A new drug that improves the sterilizing activity of current regimens would reduce the duration of therapy without sacrificing efficacy, thereby enhancing treatment completion rates and preserving precious public health resources. The new 8-methoxyfluoroquinolone moxifloxacin has potent activity against both actively multiplying and nonactively multiplying tubercle bacilli. Using a murine model that is representative of chemotherapy for human tuberculosis, we show that the combination of moxifloxacin, rifampin, and pyrazinamide reduced the time needed to eradicate Mycobacterium tuberculosis from the lungs of infected mice by up to 2 months when compared with the standard regimen of isoniazid, rifampin, and pyrazinamide. The findings suggest that this regimen has the potential to substantially shorten the duration of therapy needed to cure human tuberculosis.

The global burden of tuberculosis (TB) is staggering. An estimated 8.3 million people developed active TB and nearly 2 million people died of TB in 2000 alone (1). The World Health Organization's directly observed therapy, short-course (DOTS) strategy is an effective tool for TB control, but the implementation of DOTS has been slower than expected in developing countries, where the burden of TB is greatest (2). This is due, in part, to the inherent difficulties of administering the lengthy and labor-intensive treatment regimen in the field. In this respect, the standard short-course regimen of isoniazid (INH, H), rifampin (RIF, R), pyrazinamide (PZA, Z), and ethambutol is far from ideal. Despite its efficacy, the regimen must be administered for at least 6 months to be fully effective in humans (35). A similar duration of therapy is required to prevent relapse in the murine model (6, 7). The implementation of TB chemotherapy in the field would be much easier if the duration of therapy could be shortened without sacrificing efficacy, a feat that will require at least one new antituberculous drug (8, 9). The 8-methoxyfluoroquinolone moxifloxacin (MXF) is a new drug with promising antituberculous activity. Its MIC90 is 0.5 μg/ml for Mycobacterium tuberculosis (10). Maximum serum concentrations (Cmax) in humans are 3.2–4.5 μg/ml after a daily dose of 400 mg by the oral route and the serum half-life is 9–12 hours (1114).

MXF has bactericidal activity similar to that of INH against multiplying M. tuberculosis both in vitro and in the murine model of TB (10, 1517). It has also demonstrated early bactericidal activity approaching that of INH in patients with pulmonary TB (18, 19). However, to shorten the duration of therapy necessary to cure TB, a new drug must increase the activity of the standard regimen against persisting M. tuberculosis organisms that, by virtue of slow or intermittent multiplication, are less susceptible to killing by most antimicrobials. It is this capacity for persistence on the part of M. tuberculosis that forces therapy to be extended for many months to eradicate a small number of organisms and prevent relapse after treatment ends. The activity of MXF against “persisters” is greater than that of INH and other fluoroquinolones in an in vitro model (20); and both MXF and sparfloxacin, another fluoroquinolone with similar potency, have also demonstrated activity against persisters in the murine model of TB (6, 21, 22). Moreover, data from a clinical trial suggest that the addition of a less potent fluoroquinolone, ofloxacin, to the combination of INH, RIF, and PZA was able to shorten the duration of therapy by up to 2 months (23). We therefore hypothesized that the use of MXF would permit an even greater reduction in the duration of therapy.

Under appropriate experimental conditions, the duration of therapy required to cure tuberculosis in the murine model has been closely correlated with the duration needed to cure human tuberculosis, using the same regimen (6, 7). To address our hypothesis, we conducted an experiment in the murine model to measure the benefit of (1) adding MXF to the standard 6-month daily regimen (2 months of RIF, INH, and PZA followed with 4 months of RIF and INH [2RHZ/4RH]), or (2) substituting MXF for each first-line drug individually (Table 1)

TABLE 1. Scheme of experiment

No. of Mice Killed* at:
Treatment Group
Total No.
 of Mice
Day −18
A. Infected, untreated60151566666
B. 2RHZ/4RH426666612
C. 2RHZM/4RHM426666612
D. 2RHM/4RH426666612
E. 2RMZ/4RM426666612
F. 2MHZ/4MH426666612

*Dates are in relation to the initiation of treatment, beginning 19 days postinfection (Day 0).

Drugs were given orally five times weekly at the following doses (mg/kg): rifampin (R), 10; isoniazid (H), 25; pyrazinamide (Z), 150; moxifloxacin (M), 100. To limit pharmacologic interactions, rifampin was given alone at least 45 minutes before other drugs. Numbers before the abbreviations indicate the duration of treatment in months.

The day after infection.

§After 3 months of follow-up beyond the completion of 6 months of treatment.

. The efficacy of each regimen was assessed by (1) the time to culture-negative conversion of lung and spleen cultures, and (2) the ability to prevent culture-positive relapse after 3 months of follow-up without therapy.

Portions of the results of these studies have been previously reported in abstract form (24, 25).


MXF was provided by Bayer (Rolling Meadows, IL). INH and RIF were purchased from Sigma (St. Louis, MO) and PZA was purchased from Fisher Scientific International (Suwanee, GA). Stock solutions were prepared as previously described (15).

Mycobacterium tuberculosis Strain

Strain H37Rv of M. tuberculosis was passaged in mice to assure virulence, subcultured in Middlebrook 7H9 broth (Fisher Scientific International), and used for aerosol infection when the OD600 was 0.85. MIC values were as follows: RIF, 0.25 μg/ml; INH, 0.1 μg/ml; and MXF, 0.5 μg/ml on 7H10 medium; and PZA, 10 μg/ml on Löwenstein–Jensen medium (pH 5.5).

Aerosol Infection

Two hundred and seventy 6-week-old female BALB/c mice (Charles River Laboratories, Wilmington, MA) were infected via aerosol in a Middlebrook inhalation exposure system (Glas-Col, Terre Haute, IN). Three successive runs were performed. Mice were earmarked according to run to later confirm uniform infections with each run.


After infection, mice were randomized into six groups. Negative control mice went untreated. Positive control mice received the standard 6-month regimen, 2RHZ/4RH. A third group received the standard regimen supplemented with MXF (2RHZM/4RHM). Three other groups received the standard regimen except that MXF was substituted for PZA, INH, or RIF, respectively.

Treatment began 19 days postinfection (Day 0), and continued for 6 months. Drugs were administered by gavage five times per week in the following doses (mg/kg): RIF, 10; INH, 25; PZA, 150; MXF, 100. The MXF dose was chosen by analogy with bioequipotent doses of other fluoroquinolones (21, 26). RIF was administered at least 45 minutes before other drugs to avoid drug interactions (27, 28).

Assessment of Treatment Efficacy

Five untreated mice from each aerosol run were killed on Day –18 and on Day 0 as pretreatment controls. Six mice from each group were killed monthly during treatment. After 6 months of treatment, 8–12 mice per group went untreated for an additional 3 months and were killed to measure the culture-positive relapse rate.

Efficacy was assessed by survival rate, spleen weights, gross lung lesions (scored from + to +++, where + signified fewer than 10 tubercles per lung and +++ signified extensive tubercles), and lung and spleen colony-forming unit (CFU) counts. CFU counts were performed as previously described (15), except that plates for lung homogenates were supplemented with cycloheximide (10 μg/ml), polymyxin B (200,000 U/ml), carbenicillin (50 μg/ml), and trimethoprim (20 μg/ml) to prevent contamination (adapted from Mitchison and coworkers [29]).

Statistical Analysis

Individual CFU counts were log-transformed before analysis. Multiple pairwise comparisons of group means were performed by one-way analysis of variance with Bonferroni's posttest (30) (InStat version 3.05; GraphPad, San Diego, CA).

Rifampin and Pyrazinamide Pharmacokinetics

Uninfected 6-week-old female Swiss mice (Charles River Laboratories) were administered RHZ or RMZ under conditions similar to those of the main experiment. Solutions (R alone, HZ, and MZ) were prepared in sterile distilled water, stored overnight at 4°C, and administered in 0.2 ml by gavage. All mice received RIF first, followed 1 hour later by either HZ or MZ. At prescribed time points, mice were exsanguinated by cardiac puncture. Three samples were obtained per regimen per time point. Serum was harvested and frozen at –70°C, and then transported on dry ice to the laboratory of C. Peloquin (National Jewish Medical and Research Center, Denver, CO), where serum concentrations for RIF and PZA were determined by validated high-performance liquid chromatography assays.

Survival Rate

No mortality related to infection with M. tuberculosis occurred in untreated control mice until the third month of infection. Excluding mice killed monthly for CFU counts, one, three, one, and four control mice died in the third, fourth, fifth, and sixth months of tuberculosis infection, respectively. Therefore, 3 of 12 mice kept for mortality were still alive, although severely ill, at the end of the experiment. A total of 14 of 210 treated mice died from injuries during gavage (13) or other handling (1). Eleven of these deaths occurred within 2 weeks of the start of treatment. The group in which MXF was substituted for RIF suffered no mortality, in part because they received only one daily gavage instead of two, because RIF was administered separately from the other drugs in all other groups of mice.

Body Weight

The day before infection, the mean body weight of mice was 17.3 g. The day before onset of treatment, the mean body weight was 18.3 g, indicating that mice gained weight during the first 2 weeks of infection. During the first 2 weeks of treatment, all mice, including untreated controls, lost weight as a result of the stress of treatment and the severity of disease. Thereafter, the body weight of all treated mice increased in a similar way whatever the treatment received. The body weight of surviving untreated mice increased for the first 6 weeks after infection, reached a plateau, and then declined. At the end of the treatment phase the body weight of the three remaining untreated mice averaged 15 g, whereas that of treated mice was 21 g.

Spleen Weight

At the time of infection, the mean spleen weight (± SD) was 67 ± 5.8 mg. Eighteen days later, on Day 0, the mean spleen weight was 144.2 ± 21.2 mg. After 6 months of treatment, the spleen weights had increased to 390 ± 138 mg in untreated mice and fallen to 122 ± 27 mg in treated mice. No significant differences in spleen weight were noted between groups of treated mice.

Gross Lung Lesions

At the onset of treatment, severe lung lesions (+++) were observed in all killed mice. At 2 months, lesions were increased in number and size in untreated control mice but were reduced in size (++) in all treated mice. By the sixth month, the lungs of surviving control mice were extensively occupied by nearly coalescent tubercles. Treated mice no longer had visible lesions, except that mice in Group F had a few gross lesions (+).

Enumeration of CFU in Organs during Treatment

The day after aerosol infection (Day –18), the mean log10 CFU counts (± SD) in the lungs were 2.94 ± 0.06, 2.94 ± 0.05, and 2.90 ± 0.13 for mice infected during runs 1, 2, and 3, respectively. All spleens were culture negative. The burden of infection was therefore considered equal for mice of all three runs. At initiation of treatment (Day 0), the mean log10CFU count in the lungs had uniformly increased by 5 log10 to reach 7.80 ± 0.21. The mean spleen CFU count was 4.24 ± 0.21. In untreated control mice surviving over the entire course of treatment, the lung CFU counts remained stable between 7.2 and 8 log10 (Figure 1)

whereas the spleen CFU counts progressively increased to reach 6.77 log10 by the fourth month and then remained stable (Figure 2) .

After 2 months of treatment, lungs from mice of all treatment groups remained culture positive, but the mean CFU counts were significantly lower than pretreatment values (p < 0.01). However, the CFU counts in mice treated with the combination RMZ were dramatically lower than in mice of the reference group treated with RHZ. The mean log10 CFU counts were 0.90 ± 0.58 for the former group versus 3.36 ± 0.32 for the latter group, a difference of 2.5 log10 (p < 0.001). Figure 1 and Table 2

TABLE 2. Mean log10 colony-forming unit counts* from lung homogenates||

Duration of Treatment (months)
A. Infected, untreated7.80 ± 0.217.63 ± 0.417.24 ± 0.418.06 ± 0.817.68 ± 0.517.34 ± 0.50
B. 2RHZ/4RH3.36 ± 0.321.89 ± 0.400.39 ± 0.3200
C. 2RHZM/4RHM2.74 ± 0.481.26 ± 0.33−0.29 ± 0.5800
D. 2RHM/4RH3.70 ± 0.252.11 ± 0.261.32 ± 0.37−0.10 ± 0.57−0.65 ± 0.32§
E. 2RMZ/4RM0.90 ± 0.58−0.47 ± 0.43||000

4.21 ± 0.25
3.67 ± 0.12
3.39 ± 0.18
2.63 ± 0.35
1.98 ± 0.22

*The lowest limit of detection is a single colony out of six mice in the treatment group (i.e., –0.65 ± 0.32 log10). By convention, the value 0 denotes the absence of colonies in all mice of the treatment group.

Three mice of six yielded one to four colonies.

Four mice of six yielded one to four colonies.

§One mouse of six yielded one colony.

||Two mice of five yielded one colony.

show that the addition of MXF to RHZ (RHZM) resulted in a more modest, although significant, increase in bactericidal activity. The lung log10 CFU counts were 2.74 ± 0.48 in this group (p < 0.05 compared with positive controls). The substitution of MXF for PZA (RHM) had a small, but insignificant, impact on the CFU counts after 2 months of treatment, whereas the substitution of MXF for RIF (MHZ) had a negative impact, for the lung CFU counts were higher in mice receiving MHZ than in mice receiving RHZ (p < 0.01). The results of spleen cultures were consistent with the results in the lung, although the spleen CFU counts were only 4.24 ± 0.21 log10 at initiation of treatment. Figure 2 and Table 3

TABLE 3. Mean log10 colony-forming unit counts* from spleen homogenates

Duration of Treatment (months)
A. Infected, untreated4.24 ± 0.215.87 ± 0.186.14 ± 0.286.77 ± 0.936.37 ± 0.596.69 ± 0.67
B. 2RHZ/4RH0.04 ± 0.70−0.65 ± 0.32000
C. 2RHZM/4RHM0.40 ± 0.350000
D. 2RHM/4RH−0.23 ± 0.630000
E. 2RMZ/4RM00000

0.57 ± 0.39
0.23 ± 0.65
−0.26 ± 0.59
−0.65 ± 0.32
−0.62 ± 0.35§

*The lowest limit of detection is a single colony out of six mice in the treatment group (i.e., −0.65 ± 0.32 log10). By convention, the value 0 denotes the absence of colonies in all mice of the treatment group.

One mouse of six yielded one colony.

Three mice of six yielded one to three colonies.

§One mouse of five yielded one colony.

show that all spleen cultures were negative for mice receiving RMZ, and yielded fewer than 10 colonies for all other groups of treated mice.

After 3 months of treatment, lung cultures from mice receiving RMZ/RM were almost entirely negative: two of the five lungs (the sixth lung culture was contaminated) yielded a single CFU whereas the other three were negative. Lung cultures from mice of other groups were all positive, and the better bactericidal activity of RHZM/RHM over RHZ/RH that was observed at 2 months persisted (p < 0.01). At the following monthly sacrifices, lung cultures from mice receiving RMZ/RM became entirely negative in the fourth month, and lung cultures from mice receiving RHZ/RH and RHZM/RHM became entirely negative in the fifth month. All mice receiving RHM/RH had culture-negative lungs at 6 months, with the exception of one mouse with 1 CFU isolated. The lung cultures from all mice receiving MHZ/MH remained positive at the end of 6 months of therapy (Table 2).

The spleen cultures of mice receiving RHZM/RHM and RHM/RH became negative in the third month and remained negative in the following months, as did those of mice receiving RMZ/RM (Table 3). Among the six mice receiving RHZ/RH, only one spleen yielded a single CFU at 3 months and all spleens were culture negative thereafter. One of six mice receiving MHZ/MH had a single CFU isolated from the spleen at both the 5- and 6-month time points, meaning complete conversion to negative cultures was not obtained in this group, a finding consistent with the results of culture from the lungs of the same mice.

Enumeration of CFU in Organs after 3 Months of Follow-up: Relapse Rate

As expected from the culture results on treatment completion, all 12 lung cultures from mice that received MHZ/MH killed after 3-month follow-up without treatment were positive, with more than 104 CFU per organ. All spleen cultures but one were also positive, ranging from 50 to 5,000 CFU per organ. Only one mouse that received RHM/RH was lung culture positive, with about 500 CFU per organ, whereas its spleen culture was negative. All remaining mice that received RHM/RH as well as all mice that received RHZ/RH, RHZM/RHM, or RMZ/RM had negative lung and spleen cultures.

Serum Levels of Rifampin and Pyrazinamide

As shown in Tables 4 and 5

TABLE 4. Serum rifampin concentrations* from post hoc study

Time after Rifampin Dose (h)
RMZ5.0 ± 3.03.6 ± 3.12.2 ± 1.33.6 ± 1.04.6 ± 0.74.9 ± 1.55.7 ± 3.04.9 ± 0.8
5.0 ± 3.0
3.6 ± 3.1
1.9 ± 0.5
8.9 ± 0.4
7.7 ± 0.9
8.4 ± 3.0
4.8 ± 2.3
7.9 ± 2.5

*Micrograms per milliliter.

Drugs were given orally at the following doses (mg/kg): rifampin (R), 10; isoniazid (H), 25; pyrazinamide (Z), 150; moxifloxacin (M), 100. Rifampin was given alone 1 hour before other drugs.

TABLE 5. Serum pyrazinamide concentrations* from post hoc study

Time after Pyrazinamide Dose (h)
RMZ90 ± 1478 ± 1458 ± 1230 ± 217 ± 31.7 ± 0.4
112 ± 6
119 ± 5
92 ± 6
55 ± 7
21 ± 2
2.1 ± 0.4

*Micrograms per milliliter.

Drugs were given orally at the following doses (mg/kg): rifampin (R), 10; isoniazid (H), 25; pyrazinamide (Z), 150; moxifloxacin (M), 100. Rifampin was given alone 1 hour before other drugs.

, the serum concentrations of both RIF and PZA were not higher in mice receiving MXF instead of INH. In fact, if anything, the concentrations of PZA obtained in the RMZ group appeared lower than those obtained in the RHZ group. Thus, the better activity of the combination RMZ over the combination RHZ was not a consequence of higher concentrations of RIF or PZA.

The present experiment was designed to determine the potential benefit of using MXF as a first-line agent in combination therapy for human TB. The findings are patently clear. The addition of MXF to the standard regimen resulted in a modest, but significant, improvement in the bactericidal activity at 2 and 3 months but did not shorten the time to culture conversion by even 1 month. These experimental results support the rationale for ongoing randomized clinical trials designed to test whether the addition of MXF to the standard regimen will increase the proportion of patients with negative sputum cultures after 2 months of therapy. However, it is unclear that the addition of MXF will permit shortening of the duration of therapy. Rather, it was the substitution of MXF for INH in the standard regimen that resulted in a dramatic increase in potency. In just 2 months, the combination RMZ produced a 7-log kill in the lungs, compared with reductions of 4.5 and 5 logs with the combinations of RHZ and RHZM, respectively (p < 0.001 for both comparisons). Moreover, the lungs of mice receiving the RMZ/RM regimen had nearly complete culture conversion by 3 months of treatment, almost 2 months earlier than in mice receiving the standard regimen. Thus, it is the substitution of MXF for INH that appears to have the greatest potential to shorten the course of TB therapy. On the other hand, the substitution of MXF for RIF in the standard regimen had a negative impact. All mice treated without RIF were lung culture positive after 6 months of therapy. The substitution of MXF for PZA had little impact on the bactericidal activity of the regimen, but resulted in a more modest reduction in sterilizing activity, as culture conversion in the lungs occurred more than 1 month later in mice not receiving PZA and relapse occurred in one of eight mice, but did not occur in mice receiving the standard regimen.

The dramatic increase in potency resulting from the substitution of MXF for INH in the standard regimen is the most significant finding of the present experiment and calls for some analysis. First of all, it is not an entirely new finding. Ten years ago, Lalande and coworkers (21) demonstrated that the replacement of INH with sparfloxacin in the standard combination regimen also resulted in a more than 2 log10 reduction in CFU count after 2 months of therapy in a similar murine model. But because sparfloxacin was withdrawn from the market, further investigation was not performed.

Several explanations could be offered for the superior potency of the combination RMZ over that of RHZ, including (1) greater bactericidal activity of MXF versus INH, (2) an inappropriately high MXF dose, (3) pharmacologic interactions resulting in reduced serum concentrations of RIF and/or PZA during coadministration with INH, and (4) unanticipated synergy in the antituberculous activity of the combination of RMZ and/or unexpected antagonism with the combination of RHZ. We will consider each of these possibilities in turn.

It is unlikely that the difference in the potency of RMZ versus RHZ is a result of better individual bactericidal activity of MXF over INH. In several murine studies using the same MXF dose as in the present study, the bactericidal activity of MXF alone was not superior to that of INH (10, 15, 16). Similarly, the results of two studies comparing the early bactericidal activities of INH and MXF in patients with pulmonary TB show INH to have activity that is as good as or better than that of MXF (18, 19). Furthermore, comparison of results from groups receiving RHZ, RHZM, and RMZ clearly demonstrates that the addition of MXF to RHZ confers only modest benefit, whereas it is the removal of INH from the regimen RHZM that confers a remarkable increase in potency.

It is also unlikely that the 100-mg/kg dose of MXF was inappropriately high or otherwise nonequipotent compared with the usual human dose of 400 mg. The dose of MXF in the present study was chosen by analogy to the bioequipotent doses previously established for other fluoroquinolones (21, 26). In fact, a 100-mg/kg oral dose in mice results in an area under the serum concentration–time curve (AUC) of only 9.5 μg · hour/ml (16), whereas the 400-mg oral dose in humans results in an AUC of 30 to 40 μg · hour/ml (12, 31). Because the magnitude of the AUC-to-MIC ratio correlates with the bactericidal activity of the fluoroquinolones against other bacterial pathogens (32), the 100-mg/kg dose is more likely to underestimate rather than overestimate the activity to be expected in humans. This assumption is corroborated by pharmacodynamic data demonstrating that administration of MXF in the mouse diet at a concentration of 0.25%, which achieves both a peak serum concentration and AUC similar to that obtained with standard dosing in humans, results in greater activity than the 100-mg/kg dose by gavage (24).

A third possible explanation for the greater potency of RMZ over RHZ is a pharmacokinetic interaction resulting in different serum concentrations of RIF and/or PZA between the two treatment groups. Simultaneous administration of INH, RIF, and PZA in mice results in a 40% reduction of the Cmax of RIF, most likely because of the negative influence of INH on the absorption of RIF (27). To avoid this interaction in the present experiment, RIF was given at least 45 minutes before the simultaneous administration of the other drugs (28). We subsequently assayed serum concentrations of RIF and PZA in uninfected mice receiving RHZ or RMZ under these conditions and demonstrated that RIF and PZA concentrations were not higher in mice receiving RMZ than in mice receiving RHZ (Tables 4 and 5).

Two prior studies by Grosset and colleagues demonstrated that, when the drugs are administered to mice simultaneously, the combination RZ has significantly greater activity than the combination HRZ (27, 33). The extent to which the substantial differences in activity reflect the aforementioned pharmacokinetic interaction or reflect true differences in the antibacterial activity of the drug combinations has not been elucidated. Therefore, one could reasonably question whether the addition of MXF adds to the activity of the combination RZ at all. Previous experience with sparfloxacin suggests that a potent fluoroquinolone does add to the activity of RZ. In mice infected with M. tuberculosis by the intravenous route, the addition of sparfloxacin to a regimen of RZ resulted in an incremental reduction of more than 1 log in the lung CFU counts after 2 months of therapy (21).

In all likelihood, the dramatic increase in potency resulting from the substitution of MXF for INH in the standard regimen is the consequence of synergism in the antituberculosis activity of the three drugs RIF, MXF, and PZA and/or antagonism in that of the three drugs RIF, INH, and PZA. The mechanisms involved in the synergism and antagonism are currently unknown and deserve extensive investigation.

The final and most important point to consider is the relevance of the present findings to the clinical situation. Up to now, results obtained in the murine model have always been predictive of what was achievable in humans, provided that the relative sizes of the bacillary populations in mice and in humans are similar and the drug doses are equipotent (7). In the present experiment, as in preceding experiments (6, 34), mice responded to the standard regimen (2RHZ/4RH) much like humans taking the same drug combination. As MXF was in fact underdosed in the present experiment, there is little reason to believe that the findings observed in mice will not be observed in humans. A previous randomized human clinical trial suggested that the use of ofloxacin could shorten the duration of TB therapy by up to 2 months, although no patients were treated with a standard regimen for comparison (23). MXF has a much better pharmacodynamic profile against M. tuberculosis than ofloxacin and our results suggest that the new regimen of RIF, MXF, and PZA has the potential to shorten the duration of treatment necessary to cure human TB to 4 months or less. Further murine experiments are in progress to define more precisely the extent to which therapy can be shortened without sacrificing efficacy. These experiments will assess the relapse rate over a range of treatment durations between 3 and 6 months. Last, although a preliminary report suggests MXF will be well tolerated in humans when given for up to 6 months (35), further investigation is needed to determine the safety of MXF given in combination with RIF and PZA. Because the combination of RIF and PZA appears to have inordinate hepatotoxicity when not administered together with INH (36), one may speculate that the combination of MXF, RIF, and PZA will carry a similar risk. However, in the absence of any insights into the mechanism of PZA-induced hepatotoxicity, it is difficult to speculate further. Clearly, careful clinical investigation is warranted before considering the combination of rifampin, moxifloxacin, and pyrazinamide as a definite breakthrough in the shortening of therapy for TB.

1. Corbett EL, Watt CJ, Walker N, Maher D, Williams BG, Raviglione MC, Dye C. The growing burden of tuberculosis: global trends and interactions with the HIV epidemic. Arch Intern Med 2003;163:1009–1021.
2. Dye C, Watt CJ, Bleed D. Low access to a highly effective therapy: a challenge for international tuberculosis control. Bull World Health Organ 2002;80:437–444.
3. Fox W. Whither short-course chemotherapy? Br J Dis Chest 1981;75:331–357.
4. Blumberg HM, Burman WJ, Chaisson RE, Daley CL, Etkind SC, Friedman LN, Fujiwara P, Grzemska M, Hopewell PC, Iseman MD, et al. Treatment of tuberculosis. American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America. Am J Respir Crit Care Med 2003;167:603–662.
5. World Health Organization. Treatment of tuberculosis: guidelines for national programmes. Publication no. WHO/CDS/TB/2003.313. Geneva, Switzerland: World Health Organization; 2003.
6. Lounis N, Bentoucha A, Truffot-Pernot C, Ji B, O'Brien RJ, Vernon A, Roscigno G, Grosset J. Effectiveness of once-weekly rifapentine and moxifloxacin regimens against Mycobacterium tuberculosis in mice. Antimicrob Agents Chemother 2001;45:3482–3486.
7. Grosset J, Ji B. Experimental chemotherapy of mycobacterial diseases. In: Gangadharam PRJ, Jenkins PA, editors. Mycobacteria, Vol. II: Chemotherapy. New York: Chapman & Hall; 1998. p. 51–97.
8. O'Brien RJ, Nunn PP. The need for new drugs against tuberculosis: obstacles, opportunities, and next steps. Am J Respir Crit Care Med 2001;163:1055–1058.
9. Global Alliance for Tuberculosis Drug Development. Tuberculosis: scientific blueprint for tuberculosis drug development. Tuberculosis (Edinb) 2001;81:1–52.
10. Ji B, Lounis N, Maslo C, Truffot-Pernot C, Bonnafous P, Grosset J. In vitro and in vivo activities of moxifloxacin and clinafloxacin against Mycobacterium tuberculosis. Antimicrob Agents Chemother 1998;42:2066–2069.
11. Lubasch A, Keller I, Borner K, Koeppe P, Lode H. Comparative pharmacokinetics of ciprofloxacin, gatifloxacin, grepafloxacin, levofloxacin, trovafloxacin, and moxifloxacin after single oral administration in healthy volunteers. Antimicrob Agents Chemother 2000;44:2600–2603.
12. Sullivan JT, Woodruff M, Lettieri J, Agarwal V, Krol GJ, Leese PT, Watson S, Heller AH. Pharmacokinetics of a once-daily oral dose of moxifloxacin (Bay 12–8039), a new enantiomerically pure 8-methoxy quinolone. Antimicrob Agents Chemother 1999;43:2793–2797.
13. Stass H, Dalhoff A, Kubitza D, Schuhly U. Pharmacokinetics, safety, and tolerability of ascending single doses of moxifloxacin, a new 8-methoxy quinolone, administered to healthy subjects. Antimicrob Agents Chemother 1998;42:2060–2065.
14. Siefert HM, Domdey-Bette A, Henninger K, Hucke F, Kohlsdorfer C, Stass HH. Pharmacokinetics of the 8-methoxyquinolone, moxifloxacin: a comparison in humans and other mammalian species. J Antimicrob Chemother 1999;43(Suppl B):69–76.
15. Yoshimatsu T, Nuermberger E, Tyagi S, Chaisson R, Bishai W, Grosset J. Bactericidal activity of increasing daily and weekly doses of moxifloxacin in murine tuberculosis. Antimicrob Agents Chemother 2002;46:1875–1879.
16. Miyazaki E, Miyazaki M, Chen JM, Chaisson RE, Bishai WR. Moxifloxacin (BAY12–8039), a new 8-methoxyquinolone, is active in a mouse model of tuberculosis. Antimicrob Agents Chemother 1999;43:85–89.
17. Lenaerts AJ, Gruppo V, Brooks JV, Orme IM. Rapid in vivo screening of experimental drugs for tuberculosis using γ interferon gene-disrupted mice. Antimicrob Agents Chemother 2003;47:783–785.
18. Gosling RD, Uiso LO, Sam NE, Bongard E, Kanduma EG, Nyindo M, Morris RW, Gillespie SH. The bactericidal activity of moxifloxacin in patients with pulmonary tuberculosis. Am J Respir Crit Care Med 2003;168:1342–1345.
19. Pletz MW, Deroux A, Roth A, Mauch H, Lode H. Early bactericidal activity and efficacy of moxifloxacin versus isoniazid in the treatment of tuberculosis: a prospective, randomized study [abstract]. In: 41st Annual Meeting of the Infectious Diseases Society of America, San Diego, CA, October 9–12, 2003.
20. Hu Y, Coates AR, Mitchison DA. Sterilizing activities of fluoroquinolones against rifampin-tolerant populations of Mycobacterium tuberculosis. Antimicrob Agents Chemother 2003;47:653–657.
21. Lalande V, Truffot-Pernot C, Paccaly-Moulin A, Grosset J, Ji B. Powerful bactericidal activity of sparfloxacin (AT-4140) against Mycobacterium tuberculosis in mice. Antimicrob Agents Chemother 1993;37:407–413.
22. Veziris N, Truffot-Pernot C, Aubry A, Jarlier V, Lounis N. Fluoroquinolone-containing third-line regimen against Mycobacterium tuberculosis in vivo. Antimicrob Agents Chemother 2003;47:3117–3122.
23. Tuberculosis Research Centre C. Shortening short-course chemotherapy: a randomized clinical trial for treatment of smear positive pulmonary tuberculosis with regimens using ofloxacin in the intensive phase. Indian J Tuberc 2002;49:27–38.
24. Nuermberger E, Yoshimatsu T, Tyagi S, Bishai WR, Grosset J. Assessment of moxifloxacin activity at clinically relevant dosages in the mouse model of tuberculosis [abstract]. Am J Respir Crit Care Med 2003;169:A433.
25. Nuermberger E, Yoshimatsu T, Tyagi S, Bishai WR, Grosset J. Dramatic increase in the efficacy of combination therapy for tuberculosis (TB) with the use of moxifloxacin (M) in the murine model [abstract]. In: 43rd Annual Interscience Conference of Antimicrobial Agents and Chemotherapy, Chicago, IL, September 13–17, 2003.
26. Truffot-Pernot C, Ji B, Grosset J. Activities of pefloxacin and ofloxacin against mycobacteria: in vitro and mouse experiments. Tubercle 1991;72:57–64.
27. Grosset J, Truffot-Pernot C, Lacroix C, Ji B. Antagonism between isoniazid and the combination pyrazinamide–rifampin against tuberculosis infection in mice. Antimicrob Agents Chemother 1992;36:548–551.
28. Dhillon J, Dickinson JM, Sole K, Mitchison DA. Preventive chemotherapy of tuberculosis in Cornell model mice with combinations of rifampin, isoniazid, and pyrazinamide. Antimicrob Agents Chemother 1996;40:552–555.
29. Mitchison DA, Allen BW, Carrol L, Dickinson JM, Aber VR. A selective oleic acid albumin agar medium for tubercle bacilli. J Med Microbiol 1972;5:165–175.
30. Godfrey K. Statistics in practice: comparing the means of several groups. N Engl J Med 1985;313:1450–1456.
31. Wise R, Andrews JM, Marshall G, Hartman G. Pharmacokinetics and inflammatory-fluid penetration of moxifloxacin following oral or intravenous administration. Antimicrob Agents Chemother 1999;43:1508–1510.
32. Craig WA. Pharmacokinetic/pharmacodynamic parameters: rationale for antibacterial dosing of mice and men. Clin Infect Dis 1998;26:1–10.
33. Lecoeur HF, Truffot-Pernot C, Grosset JH. Experimental short-course preventive therapy of tuberculosis with rifampin and pyrazinamide. Am Rev Respir Dis 1989;140:1189–1193.
34. Daniel N, Lounis N, Ji B, O'Brien RJ, Vernon A, Geiter LJ, Szpytma M, Truffot-Pernot C, Hejblum G, Grosset J. Antituberculosis activity of once-weekly rifapentine-containing regimens in mice. Long-term effectiveness with 6- and 8-month treatment regimens. Am J Respir Crit Care Med 2000;161:1572–1577.
35. Valerio G, Bracciale P, Manisco V, Quitadamo M, Legari G, Bellanova S. Long-term tolerance and effectiveness of moxifloxacin therapy for tuberculosis: preliminary results. J Chemother 2003;15:66–70.
36. Centers for Disease Control and Prevention (CDC); American Thoracic Society. Update: adverse event data and revised American Thoracic Society/CDC recommendations against the use of rifampin and pyrazinamide for treatment of latent tuberculosis infection—United States, 2003. MMWR Morb Mortal Wkly Rep 2003;52:735–739.
Correspondence and requests for reprints should be addressed to Jacques H. Grosset, M.D., 1503 E. Jefferson Street, Baltimore, MD 21231-1002. E-mail:


  1. Erratum.
American Journal of Respiratory and Critical Care Medicine

Click to see any corrections or updates and to confirm this is the authentic version of record