Abstract
The objective of the present study was to compare the activities of a third-line regimen recommended by the World Health Organization (WHO) and two derivatives of that regimen with the activity of the standard combination of isoniazid, rifampin, and pyrazinamide as a positive control against Mycobacterium tuberculosis in a murine model. The WHO regimen combines ofloxacin (OFX), ethionamide, amikacin, and pyrazinamide; in the two derivatives of this regimen, OFX was replaced by levofloxacin (LVX) or moxifloxacin (MXF). The four drugs, a fluoroquinolone (either OFX, LVX, or MXF), ethionamide, pyrazinamide, and amikacin, were administered for the first 2 months (initial phase); and two drugs, a fluoroquinolone (either OFX, LVX, or MXF) and ethionamide, were administered for the following 10 months (continuation phase). After 6 months of treatment, only the spleens and lungs of mice treated with the standard regimen became culture negative. From 9 months onward, all of the organs of mice treated with the MXF-containing third-line regimen were culture negative. The majority of organs from mice treated with the OFX-containing regimen continued to be culture positive, and the mean CFU counts remained unchanged for as long as 12 months. The results for mice treated with the LVX-containing regimen fell between those for the groups receiving the MXF- and OFX-containing regimens. In conclusion, the activity of the OFX-containing third-line regimen against M. tuberculosis was rather weak in vivo, whereas when OFX was replaced by MXF, 9 months of treatment with a modified third-line regimen displayed bactericidal activity comparable to that of 6 months of treatment with the standard regimen in mice. The MXF-containing third-line regimen seems to be a powerful alternative for the treatment of tuberculosis (TB) when isoniazid and rifampin cannot be used, which is the main feature of multidrug-resistant TB.
As a consequence of the inappropriate use of essential antituberculosis (anti-TB) drugs, the bacilli of patients with TB may become resistant to one or more of these drugs. The term “multidrug-resistant (MDR) bacilli” refers to those bacilli that are resistant at least to isoniazid and rifampin, the main anti-TB drugs. Widespread dissemination of these bacilli has become a cause of grave concern for TB control in many countries (4).
Treatment of MDR TB is difficult, and the disease often carries a high rate of mortality, particularly in developing countries. Because the strains are resistant at least to isoniazid and rifampin, the therapeutic regimens must involve reserve drugs (often called “second-line” drugs), in addition to the still active essential drugs (3). Experts from the World Health Organization favor a third-line regimen containing four drugs (an aminoglycoside, ethionamide, pyrazinamide, and ofloxacin [OFX]) during the initial phase and two drugs (ethionamide and OFX) during the continuation phase (3). The regimen is recommended for patients whose drug susceptibility test results are not available, a rather common phenomenon in many developing countries. Because the effectiveness of the third-line regimen has yet to be demonstrated in either a laboratory study or a clinical trial, we have assessed its activity against Mycobacterium tuberculosis in a murine TB model. Because levofloxacin (LVX) (8, 11) and moxifloxacin (MXF) (9), two newer fluoroquinolones, have recently been demonstrated to display bactericidal activities against M. tuberculosis that are significantly greater than that of OFX, we have also compared the bactericidal activity of the third-line regimen with those of two modified regimens, in which OFX has been replaced by either LVX or MXF.
MATERIALS AND METHODS
Antimicrobial agents.
Except for isoniazid (INH), which was donated by Laphal (Allauch, France), all of the remaining compounds were purchased. Rifampin (RIF), pyrazinamide (PYZ), OFX, and LVX were manufactured by Aventis (Antony, France); amikacin (AMK) was manufactured by Bristol-Myers Squibb (La Défense, France); MXF was manufactured by Bayer (Puteaux, France); and ethionamide (ETH) was manufactured by Sigma (Saint Quentin Fallavier, France).
M. tuberculosis strain.
The H37Rv strain of M. tuberculosis was grown on Löwenstein-Jensen medium. Colonies were subcultured in Dubos broth (Diagnostics Pasteur, Paris, France) for 7 days at 37°C. The turbidity of the resulting suspension was adjusted with normal saline to match that of a standard 1-mg/ml suspension of M. bovis BCG and was further diluted with normal saline to obtain a 0.2-mg/ml suspension for inoculation into mice. The MICs of the drugs for strain H37Rv were 0.25 μg/ml for RIF, 0.06 μg/ml for INH, 1.0 μg/ml for AMK, 1.0 μg/ml for OFX, 0.5 μg/ml for LVX, 0.5 μg/ml for MXF, and 0.5 μg/ml for ETH, all of which were determined on 7H11 agar medium. The PYZ MIC for strain H37Rv was 10 μg/ml and was determined on Löwenstein-Jensen medium at pH 5.5.
Infection of mice.
Two-hundred twenty female outbred Swiss mice (age, 4 weeks) were purchased from the Janvier Breeding Center (Le Genest Saint-Isle, France) and were inoculated in the tail vein with 0.5 ml of a bacterial suspension containing approximately 1.6 × 107 CFU of M. tuberculosis H37Rv. The animal experiment guidelines of the Faculté de Médecine Pitié-Salpêtrière were followed.
Chemotherapy.
Following infection, mice were randomly allocated to five control groups (groups A to E) and three test groups (groups F to H), each with 10 to 50 mice. Group A was a negative control group, in which mice were infected but untreated; the mice in groups B, C, and D were treated with OFX, LVX, and MXF monotherapy, respectively, for 6 months; and the mice in group E, a positive control group, were treated with the standard regimen for drug-susceptible TB, i.e., 2 months of treatment with the combination INH, RIF, and PYZ, followed by 4 months of treatment with INH and RIF (10, 18). The mice in groups F, G, and H were treated for a total duration of 12 months with four drugs, i.e., a fluoroquinolone (either OFX, LVX, or MXF), ETH, AMK, and PYZ for the first 2 months, and two drugs, i.e., a fluoroquinolone and ETH, for the following 10 months. Treatment was initiated 2 weeks after infection in order to achieve a larger bacterial population and was administered 5 days per week. AMK was diluted in normal saline and given by subcutaneous injection; the remaining drugs were suspended at the desired concentrations in distilled water containing 0.05% agar and administered by gavage. Drug suspensions were prepared weekly and stored at 4°C. The drugs were administered at the following dosages per dose: RIF, 10 mg/kg of body weight; INH, 25 mg/kg; PYZ, 150 mg/kg; OFX, 200 mg/kg; LVX, 200 mg/kg; MXF, 100 mg/kg; AMK, 150 mg/kg; and ETH, 50 mg/kg. On the basis of the area under the concentration-time curves, these dosages, which are similar to those used in previous experiments (8-12, 16), were chosen to be as potent as the usual dosages administered to humans. The dosage of MXF used, i.e., 100 mg/kg, was lower than those of the other two fluoroquinolones because other investigators observed that MXF is toxic for mice when it was used at a dosage of 200 or 400 mg/kg/day (19).
Assessment of infection and treatment.
To provide baseline values, 10 and 27 untreated mice in group A were killed on days 1 and 14 after infection, respectively (days −13 and 0, respectively, in relation to the initiation of treatment). Serial killings were carried out at fixed intervals for the mice in groups treated with antibiotic combinations, whereas a single time of killing after 6 months of treatment was used for the mice in the three monotherapy groups. The severity of infection and the effectiveness of the treatments were assessed by survival rate, spleen weight, gross lung lesions (scored from 0 to ++, with the latter referring to a lung that was extensively occupied by tubercles), and the numbers of CFU in the lungs and spleens (16). At days −13 and 0 and after 2 months of treatment, the numbers of CFU in the spleens and lungs were determined by plating three serial 10-fold dilutions of homogenized suspensions onto triplicate Löwenstein-Jensen slants. After 6 months of treatment and afterward, the entire suspension prepared from each individual organ, which was supposed to contain few bacilli, was plated without dilution on 15 Löwenstein-Jensen slants. The results for the cultures were recorded after incubation at 37°C for 6 weeks. The bactericidal effect of the treatment was defined as a significant decrease in the mean number of CFU in the treated group compared with the pretreatment value.
Selection of drug-resistant mutants during treatment.
To demonstrate the selection of resistant mutants after 6 months of monotherapy with a fluoroquinolone, in addition to enumeration of the total number of CFU per organ, the undiluted organ suspensions of the mice in groups C and D were also plated onto Löwenstein-Jensen medium containing LVX at 2 μg/ml and MXF at 2 μg/ml, respectively, i.e., concentrations four times their MICs for strain H37Rv before treatment. Suspensions of the organs from mice in group B were recovered, when possible, from dead animals and were processed by the same method with Löwenstein-Jensen medium containing OFX at 2 μg/ml. Mutations leading to fluoroquinolone resistance were sought by sequencing the gyrA gene of the bacilli isolated from the fluoroquinolone-containing medium by a method described elsewhere (6).
Statistical analysis.
The results were analyzed by the Student t test and Fisher's exact probability calculation. Differences were considered significant at the 95% level of confidence.
RESULTS
Survival rate.
As expected, untreated control mice in group A began to die from day 16, and all mice in group A died by day 34 after infection. The survival rates varied widely among the treated groups.
After 1 month of treatment (corresponding to day 45 after infection), survival rates were 1 of 10, 9 of 10, and 9 of 10 for the mice in groups B, C, and D, respectively (treated with OFX, LVX, and MXF monotherapy, respectively); 19 of 20 for the mice in group E (the positive control group); and 34 of 40, 36 of 40, and 37 of 40 for the mice in groups F, G, and H, respectively (treated with the OFX-, LVX-, and MXF-containing combination regimen, respectively). The survival rate for the mice in group B did not differ significantly from that for the mice in group A (the negative control group) and was significantly lower than those for the other groups.
After 6 months of treatment, survival rates were 0 of 10, 3 of 10, and 7 of 10 for the mice in groups B, C, and D, respectively, but remained unchanged from those observed after 1 month of treatment for the mice in the remaining groups. Thus, OFX monotherapy was unable to improve the survival of M. tuberculosis-infected mice, whereas LVX or MXF monotherapy prolonged, to certain extent, the survival time; treatment with the standard regimen or any of the three fluoroquinolone-containing combination regimens effectively prevented mortality.
Mean spleen weight.
The mean spleen weight of the infected mice increased more than fourfold during the first 14 days after infection, from 111 ± 20 to 566 ± 131 mg. After 2 months of treatment, the mean spleen weights were 436 ± 70 mg for group E (the positive control group) and 363 ± 47, 386 ± 94, and 404 ± 143 mg for groups F, G, and H, respectively; all of these values were significantly less than the pretreatment weight of 566 ± 131 mg. After 4 additional months of treatment, the mean spleen weights for groups E, F, G, and H further declined significantly to 246 ± 37, 277 ± 97, 278 ± 64, and 266 ± 42 mg, respectively; the weights did not differ significantly among the groups. From 6 to 12 months of treatment, the mean spleen weights remained at levels similar to those for groups F, G, and H and were always greater than the preinfection weight of 111 ± 20 mg. These results indicate that treatment with any of the four combination regimens is equally effective in reducing the splenomegaly caused by M. tuberculosis infection.
After 6 months of monotherapy, because all of the mice had died, no spleen weight data were available for group B; the mean spleen weights of the mice in groups C and D (treated with LVX and MXF monotherapy, respectively) were 327 ± 47 and 280 ± 61 mg, respectively, and both of these were significantly less than the pretreatment weight. It should be pointed out, however, that the spleen weights for groups C and D probably represent gross underestimates, because these values do not include the spleen weights for mice that died prior to killing of the mice, which comprised 70% of the mice in group C and 30% of the mice in group D.
Gross lung lesions.
Before the start of treatment, severe (++) lung lesions were observed in all 27 control mice that were killed. After 6 months of treatment, severe (++) lung lesions remained in the majority of surviving mice in groups C and D, treated with LVX and MXF monotherapy, respectively. However, the lung lesions were quite different in groups E to H, which were treated with the combination regimens. After 2 months of treatment, the majority of mice in these groups had lung lesion scores of +, and severe (++) lung lesions were observed in only two of nine mice in group E and none of the mice in groups F, G, and H. After 6 months of treatment, lungs that appeared normal (lesion score of 0) were detected in 9 of the 10 mice in group E, 2 of the 8 mice in group F, 6 of the 9 mice in group G, and 4 of the 9 mice in group H. For those mice in groups E to H that did show lung lesions, almost all the lesions had a score of +. After 9 or 12 months of treatment, however, lung lesions were still detectable in one- to two-thirds of the mice in groups F, G and H, with the scores for almost all of them being +. These findings indicate that 2 months of combination therapy significantly alleviated the lung lesions caused by M. tuberculosis infection and that 4 additional months of treatment resulted in further amelioration, but minor lesions remained in a proportion of mice even after 9 or 12 months of treatment. The degree of improvement of the gross lung lesions did not differ significantly among mice treated with the four different combination regimens.
Enumeration of CFU in organs.
Except for the mice in groups A and B, all of which had died during the first 2 months of the experiment, the results of culture of the organs are summarized in Table 1. As demonstrated in previous experiments (5), by the time that the mice died from M. tuberculosis infection, their spleens and lungs contained, on average, 107 to 108 bacilli per organ, suggesting that OFX monotherapy in group B failed to prevent multiplication of the bacilli. Except for the lungs of mice in group C (treated with LVX monotherapy for 6 months), all of the values in Table 1 were significantly less than the pretreatment values.
TABLE 1.
Organism counts per organ after various durations of treatment
| Groupa | Regimenb | Mean no. (log10) of CFU/organ (no. of culture-positive organs/no. of organs tested)
|
|||||||
|---|---|---|---|---|---|---|---|---|---|
| Spleensc
|
Lungsc
|
||||||||
| 2 mo | 6 mo | 9 mo | 12 mo | 2 mo | 6 mo | 9 mo | 12 mo | ||
| C | 6 mo of LVX alone | 4.51 ± 0.42 (3/3) | 6.66 ± 0.79d (3/3) | ||||||
| D | 6 mo of MXF alone | 4.28 ± 0.36 (7/7) | 5.32 ± 0.45 (7/7) | ||||||
| E | 2 mo of RIF-INH-PYZ + 4 mo of RIF-INH | 2.49 ± 0.48 (9/9) | <0.07g,h,i (0/10) | 3.21 ± 0.54 (9/9) | <0.07i (0/10) | ||||
| F | 2 mo of OFX-ETH-PYZ-AMK + 10 mo of OFX-ETH | 2.80 ± 0.33 (8/8) | 1.50 ± 0.27i (9/9)j | 1.27 ± 0.80 (6/8)j | 1.37 ± 1.26 (7/8)j | 3.45 ± 0.34 (8/8) | 2.15 ± 0.45i (9/9)j | 1.34 ± 0.67 (7/8)j | 1.71 ± 1.89 (7/8)j,k |
| G | 2 mo of LVX-ETH-PYZ-AMK + 10 mo of LVX-ETH | 2.84 ± 0.32 (9/9) | 1.07 ± 0.34g,i (9/9)j | 0.74 ± 0.64 (9/10)j | 0.10 ± 0.15 (5/9) | 3.38 ± 0.22 (9/9) | 0.91 ± 0.63g,i (8/9)j | 0.29 ± 0.36g,i (7/10)j | 0.07e,g (1/9)l |
| H | 2 mo of MXF-ETH-PYZ-AMK + 10 mo of MXF-ETH | 2.09 ± 0.39g,h (9/9) | 0.15 ± 0.18g,h,i (4/9) | <0.07 (0/10) | <0.07 (0/9) | 1.82 ± 0.39f,g,h (9/9) | <0.07i (0/9)l | <0.07g (0/10) | <0.07g (0/9) |
No CFU data were available at 2 months or later for groups A (negative control) and B (treated with OFX alone for 6 months) because all mice had died.
Treatment began 14 days after infection and was administered five times weekly. The drugs were given at the dosages described in the text. Except AMK was given by subcutaneous injection; all other drugs were given by esophageal gavage.
Mice were infected intravenously with 1.67 × 107 CFU of M. tuberculosis H37Rv. On the day after infection the mean number (log10) of CFU in the spleens and lungs of 10 mice were 5.49 ± 30 and 5.12 ± 0.23, respectively; 2 weeks later, by the time that treatment began, the numbers in the spleens and lungs in 27 mice increased to 7.28 ± 0.42 and 7.86 ± 0.25, respectively.
The only value in the table that does not differ significantly from the pretreatment (day 0) value.
A single colony was observed on only 1 of the 15 Löwenstein-Jensen slants which had been plated with 0.2 ml of undiluted organ suspension. The value also represents the lower limit of detectability of the CFU count for an organ.
Significantly smaller than the mean number of CFU for mice in group E killed simultaneously.
Significantly smaller than the mean number of CFU for mice in group F killed simultaneously.
Significantly smaller than the mean number of CFU for mice in group G killed simultaneously.
Significantly smaller than the mean number of CFU for mice in the same group killed earlier.
Proportion of culture positivity significantly greater than that for mice in group H killed simultaneously.
Proportion of culture positivity significantly greater than that for mice in group G killed simultaneously.
Proportion of culture positivity significantly smaller than that for mice in the same group killed earlier.
After 2 months of treatment, although all of the organs of the treated mice remained culture positive, the numbers of CFU in either the spleens or the lungs of the mice in groups E to H had been reduced at least 4 log10 from their pretreatment values. The greatest reduction occurred in group H, in which reductions of more than 5 log10 CFU in the spleens and 6 log10 CFU in the lungs were observed; consequently, the numbers of CFU in the organs of mice in this group were significantly smaller than the corresponding numbers in the organs of mice in groups E, F, and G. The numbers of CFU did not differ significantly among the last three groups.
After 6 months of treatment, the numbers of CFU in the mice in any of the four groups treated with a combination regimen were significantly smaller than those in mice treated with monotherapy. It appears likely that the values in Table 1 for mice in groups C and D represent underestimates, because the tabulated values do not include the counts for organs from mice that died prior to 6 months (7 of 10 mice in group C and 3 of 10 mice in group D). At this point, differences between groups E and H, on the one hand, and groups F and G, on the other, became apparent. Although 4 additional months of treatment further reduced significantly the numbers of CFU in both organs of the mice in groups F and G, all of the organs remained culture positive; however, all of the organs from mice in group E and all of the lungs from mice in group H had become culture negative. The proportion of culture-positive spleens and the mean number of CFU in the spleens of mice group H were significantly smaller than those for groups F and G.
After 9 months of treatment, whereas all the organs from mice in group H had become culture negative, the majority of the organs from mice in groups F and G remained culture positive. Compared with the values at 6 months, the numbers of CFU in the mice in group F remained unchanged, but they were reduced significantly for mice in group G.
After 12 months of treatment, whereas all the organs from mice in group H remained culture negative, the majority of organs from mice in group F were still culture positive and the numbers of CFU were basically unchanged from those at 9 months. For mice in group G, the proportion of culture-positive organs (five of nine spleens and one of nine lungs) and the numbers of CFU were reduced from those at 9 months. The mean number of CFU in the lungs was at the lower limit of detectability and was similar to that found after 6 months for mice in group H.
Selection of fluoroquinolone-resistant mutants during treatment.
After 6 months of treatment, mutants resistant to the fluoroquinolone used for treatment were isolated from the spleens and lungs of all 10 surviving mice in groups C and D, treated with LVX and MXF monotherapy, respectively. On the basis of a comparison of the numbers of CFU of fluoroquinolone-resistant mutants and the total CFU counts in the organs, almost all the bacilli recovered from the spleens and lungs of mice in groups C and D were fluoroquinolone resistant after 6 months of treatment (means, 66 to 98% for group C and 76 to 85% for group D). In contrast, the proportion of resistant mutants present in the organs of mice in group B by the time that the mice died (during the first month of treatment) was about 10−6 of the bacterial population, i.e., close to the natural proportion of mutants of strain H37Rv described in the literature (15) and found in our experiment in the organs of control mice at the start of treatment.
The mutations in the quinolone resistance-determining region (2) of the gyrA gene found in 9 of 10 resistant clones isolated from mice in groups C and D involved substitution of amino acid position 83 or 87, the most commonly substituted amino acid found in fluoroquinolone-resistant mycobacteria (6).
DISCUSSION
The objective of the experiment described here was to compare the activities of the recommended third-line regimen and its two modifications with that of the standard combination of RIF, INH, and PYZ against M. tuberculosis in the murine model. It was recommended that, for the treatment of human TB, the duration of the initial phase of the regimen be 3 months and that the duration of the continuation phase of the regimen be 18 months (3). Because of the limited life span of the mouse, however, we had to reduce the duration of drug administration to 2 months for the initial phase and to 10 months for the continuation phase in the experiments with mice.
The regimen administered to the mice in group F was that recommended by the World Health Organization, i.e., AMK-ETH-OFX-PYZ. The results demonstrated that at the end of the initial phase of treatment, both the proportion of culture-positive organs and the numbers of CFU for mice in group F did not differ significantly from those for mice in group E, which were treated with RIF-INH-PYZ, the standard regimen, indicating that during the initial phase the OFX-ETH-PYZ-AMK third-line regimen displays bactericidal activity comparable to that of the standard combination regimen (RIF-INH-PYZ). Once the initial phase shifted to the continuation phase, however, the bactericidal activities of the two regimens differed significantly. After 6 months of treatment, while all the organs of the mice in group E were culture negative, all of those of the mice in group F were culture positive. Furthermore, the organ culture positivity and the numbers of CFU in group F remained unchanged from 6 to 12 months of treatment, indicating that treatment with OFX-ETH for 10 months was significantly less effective than treatment with RIF-INH for 4 months. Because the in vivo activity of the third-line regimen with OFX as the fluoroquinolone against M. tuberculosis was rather weak, the prospect for the treatment of MDR TB with this regimen is bleak.
Because our previous experiments have demonstrated that OFX displayed only modest bactericidal activity against M. tuberculosis (8, 9, 11, 16), one of the possible approaches to improve the effectiveness of the third-line regimen is to replace OFX with another fluoroquinolone that has been demonstrated to possess more powerful bactericidal activity against M. tuberculosis. On the basis of earlier observations that LVX is approximately twice as active against M. tuberculosis as OFX both in vitro and in vivo (8, 11), the mice in group G were administered a modified third-line regimen, in which OFX was replaced by LVX. As shown in Table 1, during the initial phase the bactericidal activity of LVX-ETH-PYZ-AMK did not differ significantly from that of either RIF-INH-PYZ or OFX-ETH-PYZ-AMK; during the continuation phase, however, although the proportion of culture-positive organs and the numbers of CFU steadily declined in the mice in group G and were smaller than the corresponding values for the mice in group F after 12 months of treatment, neither the spleens nor the lungs of the mice in group G ever became 100% culture negative in the course of the experiment. These results suggest that LVX-ETH is marginally more active than OFX-ETH. In addition, bactericidal activity after 10 months of the continuation phase with LVX-ETH failed to match that after 4 months of the continuation phase with RIF-INH. It is therefore unlikely that the LVX-based modified third-line regimen will be sufficiently effective for the treatment of MDR TB.
MXF is the fluoroquinolone that is by far the most bactericidal against M. tuberculosis in the murine TB model (1, 9, 14, 19). Recent findings also suggest that MXF exhibits sterilizing activity against M. tuberculosis, including RIF-tolerant populations, which might make it possible to shorten the duration of treatment (7, 12). The mice in group H were administered another modified third-line regimen in which OFX was replaced by MXF. At the end of the initial phase, the numbers of CFU in the mice in group H were significantly smaller than those in mice treated with any of the three other combination regimens, including the standard regimen, indicating that 2 months of treatment with MXF-ETH-PYZ-AMK was significantly more bactericidal than treatment with RIF-INH-PYZ for the same duration. However, during the continuation phase, although the reduction in the numbers of CFU was far more rapid in the mice in group H than the mice in groups F and G, it was slower than that in the mice in group E. Only 4 months of treatment with RIF-INH was required to achieve culture negativity for all organs, whereas 7 months of treatment with MXF-ETH was required to achieve culture negativity. Nonetheless, the MXF-based modified third-line regimen appears to represent the first regimen not containing RIF or INH, the two major antituberculous agents, capable of rendering all the organs culture negative after only 9 months of treatment. This MXF-based third-line regimen is a powerful alternative regimen for the treatment of MDR TB.
One of the major aims of chemotherapy is to prevent the emergence of drug resistance, which is particularly important during treatment for MDR TB. After 6 months of monotherapy, fluoroquinolone-resistant mutants were isolated from all surviving mice that had been treated with either LVX or MXF; moreover, the large majority of the bacilli recovered from the organs of these mice were fluoroquinolone resistant. All mice that had been treated with OFX monotherapy at a dosage of 200 mg/kg 5 times weekly died much earlier (i.e., during the first month) and were infected with susceptible bacilli at the time of death (i.e., the bacillus population contained the same proportion of mutants [about 10−6] as the wild-type strain). These data represent additional evidence of the higher levels of activity of LVX and MXF compared with that OFX, at least at the dosages used in the experiments described here.
The results for the mice in groups G and H after 12 months of treatment indicate that the drugs which accompanied the fluoroquinolone had successfully prevented the selection of fluoroquinolone-resistant mutants, since there were virtually no surviving bacilli at the end of treatment.
The results of the present experiments demonstrate that during the initial 2 months of treatment, the four-drug combination of the third-line regimen that includes a fluoroquinolone, particularly MXF, displayed bactericidal activity at least comparable to that of RIF-INH-PYZ, the standard regimen. The main weakness of the third-line regimens appears to reside in the two-drug combinations used during the continuation phase, even for the MXF-containing regimen, and none of them exhibits bactericidal activity comparable to that of RIF-INH. Therefore, to improve further the effectiveness of the third-line regimen, its composition during the continuation phase must be strengthened. For that there appear to be two possible options: first, to replace ETH by another drug that shows more powerful sterilizing activity (7, 13) and that is better tolerated, and second, to add a third bactericidal drug. At present, however, there is no suitable candidate that can replace or supplement ETH for the treatment of MDR TB. Consequently, screening for drugs with more potent sterilizing activities against M. tuberculosis must be continued.
The efficiency of the MXF-containing third-line regimen for the treatment of human MDR TB must be validated in clinical trials. Furthermore, because no information is available about adverse reactions during long durations of daily MXF treatment and because hepatotoxicity was reported to be frequently associated with administration of the combination of ETH and PYZ (17), the tolerance of the modified third-line regimen by patients must be closely monitored.
Acknowledgments
We thank Jérome Robert for statistical assistance and Murielle Renard for technical assistance.
This study was partially funded by a grants from Ministère de l'Education Nationale et de la Recherche (grant UPRES 1541) and the Institut National de la Santé et de la Recherche Médicale (grant EMI 0004).
REFERENCES
- 1.Alvirez-Freites, E. J., J. A. Carter, and M. H. Cynamon. 2002. In vitro and in vivo activities of gatifloxacin against Mycobacterium tuberculosis. Antimicrob. Agents Chemother. 46:1022-1025. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Cambau, E., W. Sougakoff, M. Besson, C. Truffot-Pernot, J. Grosset, and V. Jarlier. 1994. Selection of a gyrA mutant of Mycobacterium tuberculosis resistant to fluoroquinolones during treatment with ofloxacin. J. Infect. Dis. 170:479-483. [DOI] [PubMed] [Google Scholar]
- 3.Crofton, J., P. Chaulet, and D. Maher. 1997. Guidelines for the management of drug-resistant tuberculosis. Report WHO/TB/96.210. World Health Organization, Geneva, Switzerland.
- 4.Espinal, M. A., A. Laszlo, L. Simonsen, F. Boulahbal, S. J. Kim, A. Reniero, S. Hoffner, H. L. Rieder, N. Binkin, C. Dye, R. Williams, and M. C. Raviglione. 2001. Global trends in resistance to antituberculosis drugs. N. Engl. J. Med. 344:1294-1303. [DOI] [PubMed] [Google Scholar]
- 5.Grosset, J., and B. Ji. 1998. Experimental chemotherapy of mycobacterial diseases, p. 51-97. In P. R. J. Gandharam and P. A. Jenkins (ed.), Mycobacteria, vol. II. Chemotherapy. Chapman & Hall, New York, N.Y.
- 6.Guillemin, I., V. Jarlier, and E. Cambau. 1998. Correlation between quinolone susceptibility patterns and sequences in the A and B subunits of DNA gyrase in mycobacteria. Antimicrob. Agents Chemother. 42:2084-2088. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hu, Y., A. R. M. Coates, and D. A. Mitchison. 2003. Sterilizing activities of fluoroquinolones against rifampin-tolerant populations of Mycobacterium tuberculosis. Antimicrob. Agents Chemother. 47:653-657. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Ji, B., N. Lounis, C. Truffot-Pernot, and J. Grosset. 1995. In vitro and in vivo activities of levofloxacin against Mycobacterium tuberculosis. Antimicrob. Agents Chemother. 39:1341-1344. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Ji, B., N. Lounis, C. Maslo, C. Truffot-Pernot, P. Bonnafous, and J. Grosset. 1998. In vitro and in vivo activities of moxifloxacin and clinafloxacin against Mycobacterium tuberculosis. Antimicrob. Agents Chemother. 42:2066-2069. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Lecoeur, H. F., C. Truffot-Pernot, and J. H. Grosset. 1989. Experimental short-course preventive therapy of tuberculosis with rifampin and pyrazinamide. Am. Rev. Respir. Dis. 140:1189-1193. [DOI] [PubMed] [Google Scholar]
- 11.Lounis, N., B. Ji, C. Truffot-Pernot, and J. Grosset. 1997. Which aminoglycoside or fluoroquinolone is more active against Mycobacterium tuberculosis in mice? Antimicrob. Agents Chemother. 41:607-610. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Lounis, N., A. Bentoucha, C. Truffot-Pernot, B. Ji, R. J. O'Brien, A. Vernon, G. Roscigno, and J. Grosset. 2001. Effectiveness of once-weekly rifapentine and moxifloxacin regimens against Mycobacterium tuberculosis in mice. Antimicrob. Agents Chemother. 45:3482-3486. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Mitchison, D. A. 2000. Role of individual drugs in the chemotherapy of tuberculosis. Int. J. Tuberc. Lung Dis. 4:796-806. [PubMed] [Google Scholar]
- 14.Miyazaki, E., M. Miyazaki, J. M. Chen, R. E. Chaisson, and W. R. Bishai. 1999. Moxifloxacin (BAY 12-8039), a new 8-methoxyquinolone, is active in a mouse model of tuberculosis. Antimicrob. Agents Chemother. 43:85-89. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Takiff, H. E., L. Salazar, C. Guerrero, W. Philipp, W. M. Huang, B. Kreiswirth, S. T. Cole, W. R. Jacobs, Jr., and A. Telenti. 1994. Cloning and nucleotide sequence of Mycobacterium tuberculosis gyrA and gyrB genes and detection of quinolone resistance mutations. Antimicrob. Agents Chemother. 38:773-780. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Truffot-Pernot, C., B. Ji, and J. Grosset. 1991. Activities of pefloxacin and ofloxacin against mycobacteria: in vitro and mouse experiments. Tubercle 72:57-64. [DOI] [PubMed] [Google Scholar]
- 17.Wada, M. 1998. The adverse reactions of anti-tuberculosis drugs and its management. Nippon Rinsho 56:3091-3095. [PubMed] [Google Scholar]
- 18.World Health Organization. 1997. Treatment of tuberculosis: guidelines for national programmes, 2nd ed. Report WHO/TB/97.220. World Health Organization, Geneva, Switzerland.
- 19.Yoshimatsu, T., E. Nuermberger, S. Tyagi, R. Chaisson, W. Bishai, and J. Grosset. 2002. Bactericidal activity of increasing daily and weekly doses of moxifloxacin in murine tuberculosis. Antimicrob. Agents Chemother. 46:1875-1879. [DOI] [PMC free article] [PubMed] [Google Scholar]