Activities of Linezolid against Rapidly Growing Mycobacteria

Abstract

Linezolid is an oxazolidinone available as an oral drug which has activity against most gram-positive bacteria. However, few species of the genus Mycobacterium have been studied. We tested 249 clinical isolates and 10 reference strains of rapidly growing mycobacteria for susceptibility to linezolid by broth microdilution. Clinical species included the Mycobacterium fortuitum group (n = 74), M. abscessus (n = 98), M. chelonae (n = 50), M. mucogenicum (n = 10), and M. fortuitum third biovariant complex (10). The modal MIC for M. mucogenicum was 1.0 μg/ml, and the MIC at which 90% of the isolates tested are inhibited (MIC₉₀) was 4 μg/ml; the modal MIC for the M. fortuitum group was 4 μg/ml, and the MIC₉₀ was 16 μg/ml; the modal MIC for the M. fortuitum third biovariant complex was 4 μg/ml, and the MIC₉₀ was 8 μg/ml; the modal MIC for M. chelonae was 8 μg/ml, and the MIC₉₀ was 16 μg/ml; and the modal MIC for M. abscessus was 32 μg/ml, and the MIC₉₀ was 64 μg/ml. Based on peak levels of linezolid in serum of 15 to 20 μg/ml, we propose the following broth MIC breakpoints for these species: susceptible, ≤ 8 μg/ml; moderately susceptible, 16 μg/ml; and resistant, ≥32 μg/ml). These studies demonstrate the excellent potential of linezolid for therapy of rapidly growing mycobacteria.

Treatment of infections due to nontuberculous mycobacteria remains difficult, in part because they are resistant to many of the first-line tuberculosis agents and in part because so few other agents are available for therapy (21). The oxazolidinones are one of several new classes of agents active against gram-positive bacteria (4, 8, 11; M. C. Birmingham, G. S. Zimmer, B. Hafkin, W. M. Todd, T. Leach, D. H. Batts, S. M. Flavin, C. R. Rayner, K. E. Welch, P. F. Smith, J. D. Root, N. E. Wilks, and J. J. Schentag, Abstr. 39th Intersci. Conf. Antimicrob. Agents Chemother., abstr. 1098, 1999) which have the potential for activity against nontuberculous mycobacteria, including the rapidly growing mycobacteria (6; M. Wu, P. Aralor, K. Nash, L. E. Bermudez, C. B. Inderlied, and L. S. Young, Abstr. 38th Intersci. Conf. Antimicrob. Agents Chemother., abstr. E-143, 1998; L. E. Bermudez, Abstr. Int. Conf. Macrolides, Azalides, Streptogramins, Ketolides and Oxazolidinone, abstr. 03.16, 2000).

We chose to study linezolid, which along with eperezolid is the first of the oxazolidinones to reach clinical testing (4; M. C. Birmingham et al., 39th ICAAC) Linezolid offers special promise, as it is a twice-daily oral drug which is 100% bioavailable (Zyvox package insert, 2000 [Pharmacia & Upjohn, Inc.]; also see reference 8) and appears to be well tolerated (8; N. E. Wilks, Abstr. 39th Intersci. Conf. Antimicrob. Agents Chemother., abstr. 1763, 1999), important features of therapeutic drugs to be used against nontuberculous mycobacteria which cause chronic infections and require long-term therapy (often 6 months or longer) (21). The activities of linezolid against a large number of clinical isolates of rapidly growing mycobacteria, including the common disease-producing species Mycobacterium fortuitum, M. chelonae, and M. abscessus (24), were studied.

(This work was presented in part as an abstract at the 100th General Meeting of the American Society for Microbiology, Los Angeles, Calif.)

MATERIALS AND METHODS

Organisms.

We tested 249 clinical isolates of rapidly growing mycobacteria belonging to seven taxonomic groups which had been submitted for susceptibility testing to the Mycobacteria/Nocardia Laboratory at the University of Texas Health Center between 1998 and 2000. Organisms were identified to the species level by PCR restriction fragment length polymorphism analysis of the 439-bp (Telenti) segment of the 65-kDa hsp gene (14, 17) or by pattern of drug susceptibility to approximately 15 other drugs, including aminoglycosides, beta-lactams, and quinolones (2). The test species used and the number of clinical test isolates (in parentheses) were as follows: M. fortuitum group without separation into M. fortuitum, M. peregrinum, or M. fortuitum third biovariant complex (n = 74); M. fortuitum third biovariant complex (n = 10 [with five each being sorbitol positive and sorbitol negative]); M. abscessus (n = 98); M. chelonae (n = 50); M. mucogenicum (n=10); M. immunogenum (n=4) (see reference 25; and M. smegmatis group (n=3). Four reference strains obtained or submitted by our laboratory to the American Type Culture Collection (Manassas, Va.) were also tested. These included M. peregrinum ATCC 700686, M. fortuitum ATCC 6841^T, M. chelonae ATCC 35752^T, and M. abscessus ATCC 19977^T.

Susceptibility testing.

Susceptibility testing utilized serial twofold broth microdilution in cation supplemented Mueller-Hinton broth as previously described (2). Susceptibilities to linezolid were read after incubation at 30°C in room air for 3 days. The endpoint was complete (100%) inhibition of visible growth.

Quality control.

Quality control was performed using Staphylococcus aureus ATCC 29213. The acceptable range of inhibition for this strain was 1 to 4 μg/ml (18 to 24 h of incubation) (linezolid package insert, 2000 [Pharmacia & Upjohn, Inc.]). M. peregrinum ATCC 700686 was also tested as a rapid-grower control as recently recommended (26). Both were used initially after preparation of the MIC panels and then weekly.

RESULTS

Susceptibility testing.

Susceptibility to linezolid was determined for 249 clinical isolates of rapidly growing mycobacteria belonging to seven taxonomic groups. In general, a unimodal distribution of MICs was seen for the common species or groups, with >90% of MICs within two dilutions of the mode. Of the common pathogenic species, the drug was most active against isolates of M. mucogenicum and the M. fortuitum group, including both members of M. fortuitum third biovariant complex. Of the M. mucogenicum isolates, 6 of 10 (60%) were inhibited by linezolid at concentrations of 1 μg/ml, and 100% were inhibited by linezolid at 8 μg/ml. For M. fortuitum, 45 of 74 (61%) of the test isolates were inhibited by linezolid at concentrations of 4 μg/ml or less, and 96% were inhibited by linezolid at concentrations 16 μg/ml or less, with a mode of 4 μg/ml. For M. fortuitum third biovariant complex, all 10 isolates were within one dilution of the mode of 4 μg/ml (Tables 1 and 2).

TABLE 1.

Susceptibility of 249 clinical isolates of seven species or taxa of rapidly growing mycobacteria to linezolid as determined by broth microdilution

Species	No. (cumulative %) of isolates inhibited by MIC (μg/ml) of:										No. of isolates tested
	0.25	≤0.5	1	2	4	8	16	32	64	≥128
M. fortuitum groupa	0 (0)	0 (0)	4 (5)	15 (26)	26 (61)	19 (86)	7 (96)	3 (100)	0 (100)	0	74
M. fortuitum third biovariant complex	0 (0)	0 (0)	0 (0)	1 (10)	6 (70)	3 (100)	0	0	0	0	10
M. abscessus	0 (0)	2 (2)	0 (2)	1 (3)	6 (9)	14 (23)	24 (48)	39 (88)	10 (98)	2 (100)	98
M. chelonae	0 (0)	0 (0)	1 (2)	0 (0)	5 (12)	21 (54)	20 (94)	2 (98)	1 (100)	0 (100)	50
M. mucogenicum	0 (0)	2 (20)	4 (60)	1 (70)	2 (90)	1 (100)	0	0	0	0	10
M. smegmatis groupb	0 (0)	1 (33)	0 (33)	1 (67)	1 (100)	0	0	0	0	0	3
M. immunogenum	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	2 (50)	2 (100)	0	0	4

^aIncludes M. fortuitum, M. peregrinum, and M. fortuitum third biovariant complex. 

^bIncludes M. goodii and M. smegmatis sensu stricto. 

TABLE 2.

Susceptibility of 249 isolates of rapidly growing mycobacteria to linezolid as determined by broth microdilution

Organism	No. of isolates tested	MIC (μg/ml)c				% Susceptible or intermediate (MIC ≤16 μg/ml)
		Range	50%	90%	Mode
M. fortuitum groupa	74	1–32	4	16	4	96
M. fortuitum third biovariant complex	10	2–8		8		100
M. abscessus	98	0.5–128	32	64	32	48
M. chelonae	50	1–64	8	16	8	94
M. mucogenicum	10	0.5–8		4		100
M. smegmatis groupb	3	0.5–4		4		100
M. immunogenum	4	16–32		32		50
Range or total	249	0.5–128	1–32	4–64	0.5–32	48–100

^aIncludes M. fortuitum, M. peregrinum, and M. fortuitum third biovariant complex. 

^bIncludes M. goodii and M. smegmatis sensu stricto. 

^c50% and 90%, MICs at which 50 and 90% of isolates are inhibited, respectively. 

The drug was also active when tested against isolates of M. chelonae. Twenty-seven of 50 (54%) isolates were inhibited by linezolid at concentrations of 8 μg/ml or less, and 47 of 50 (94%) were inhibited by linezolid at 16 μg/ml or less. The mode for M. chelonae was 8 μg/ml, with 41 of 50 isolates (82%) having an MIC of either 8 or 16 μg/ml (Tables 1 and 2).

Isolates of M. abscessus were the least susceptible to linezolid of the three common species of rapidly growing mycobacteria. With testing of 98 isolates and 100% growth inhibition, only 23 of 98 (23%) isolates were inhibited by linezolid at concentrations of 8 μg/ml or less, and only 47 of 98 (48%) were inhibited by linezolid at 16 μg/ml (Tables 1 and 2).

The linezolid MIC for M. fortuitum ATCC 6841^T was 8 μg/ml, that for M. chelonae ATCC 35752^T was 1 μg/ml, and that for M. abscessus ATCC 19977^T was 64 μg/ml.

Quality control.

The S. aureus control strain was within the expected MIC range. The strain of M. peregrinum ATCC 700686 recommended as an MIC control strain (26) was tested 54 times. The modal MIC was 4μg/ml (31 values), with 11 values at 2 μg/ml and 9 values at 8 μg/ml. A total of 51 of 54 (94%) values were within one dilution of the mode, and 100% were within two dilutions.

DISCUSSION

The manufacturer of linezolid has proposed ≥8 μg/ml as the resistance breakpoint of linezolid for Enterococcus spp. (package insert for Zyvox, 2000 [Pharmacia & Upjohn Inc.]). For Staphylococcus species the susceptibility breakpoint is ≤4 μg/ml, and for Streptococcus spp. it is ≤2 μg/ml. No resistance breakpoints were given for these two groups because of the absence of isolates for which the linezolid MICs were >4 μg/ml and hence no experience with isolates such as these rapidly growing mycobacteria which have higher MICs. Peak levels of the drug in serum after oral doses of 600 mg twice a day (mean ± standard deviation are 21.2 ± 5.8 μg/ml, with a half-life of 5.4 h (package insert for Zyvox, 2000 [Pharmacia & Upjohn Inc.]), suggesting that 16 μg/ml might be an acceptable intermediate value for other organisms. Pending clinical experience, we propose the following breakpoints for these mycobacterial species: susceptible, ≤8 μg/ml; intermediate, 16 μg/ml; and resistant, ≥32 μg/ml.

Drug treatment of rapidly growing mycobacteria has been severely limited by the small number of available drugs with activity at clinically achievable levels in tissue (blood), especially oral drugs (15, 21, 24). For M. chelonae and M. abscessus, clarithromycin (and the other macrolides) is the only oral agent active against all untreated strains (3, 15, 18, 19, 22). Acquired mutational resistance is a concern with monotherapy with this drug, and several mutations have been described which were associated with marked increases in macrolide MICs and clinical treatment failure (16, 22). Linezolid offers great promise as a macrolide companion drug for isolates of M. chelonae, as 94% of all isolates tested were inhibited by this drug at concentrations of ≤16 μg/ml, with a mode of 8 μg/ml, including strains with acquired clarithromycin resistance (data not shown).

The MICs were higher for M. abscessus than for M. chelonae, but 48% of these isolates were inhibited by linezolid at concentrations of 16 μg/ml. Chronic lung disease in the setting of bronchiectasis (nodular bronchiectasis in elderly women or young adolescents with cystic fibrosis) is one of the more common forms of clinical disease due to M. abscessus (5, 9, 10, 12, 24) and is currently incurable with drugs alone (5, 10, 12, 21). Despite these MICs, linezolid is one of the few drugs to offer promise for treatment of this disease.

The third common pathogenic rapidly growing mycobacterial species (group) to be tested was the M. fortuitum group. This group includes M. fortuitum, M. peregrinum, M. fortuitum third biovariant (sorbitol positive), and M. fortuitum third biovariant (sorbitol negative) (23). These species and taxa have very similar drug susceptibilities (15, 18, 20). Most or all untreated strains are susceptible to the newer quinolones and sulfamethoxazole (15, 18), although acquired mutational resistance to the newer quinolones (18) limits the use of this class of drugs as monotherapy. Some but not all isolates of M. fortuitum are also susceptible to doxycycline and minocycline (42%) (15) and clarithromycin (3). Isolates of M. peregrinum and M. fortuitum third biovariant sorbitol negative are all susceptible to clarithromycin (MIC, ≤4 μg/ml) while all isolates of M. fortuitum third biovariant (sorbitol positive) are clarithromycin resistant (MICs > 4 μg/ml) (3). For the M. fortuitum group, 61% of isolates were inhibited by 4 μg of linezolid per ml, and 86% were inhibited by 8 μg of linezolid per ml. Thus, linezolid offers the promise of an additional oral drug which could be used for all isolates of this species or group.

The MIC endpoint used in this study was 100% inhibition of visible growth, an endpoint suggested by Pharmacia & Upjohn, Inc. Instructions with the recently released linezolid E-test (AB Biodisk), however, are to use an 80% growth inhibition endpoint rather than 100%. E-test MIC data for clinical isolates in the present study were not available because the 80% growth inhibition endpoint was not studied. A trailing endpoint appeared to be minimal with these groups or species by broth microdilution, as use of 80% inhibition compared to 100% inhibition only decreased the MICs by an average of one dilution (data not shown). Only clinical trials will determine if the suggested broth MIC breakpoints for these mycobacteria (susceptible ≤8 μg/ml, moderately susceptible = 16 μg/ml, resistance ≥32 μg/ml) is appropriate or correct.

Acquired resistance to linezolid among gram-positive bacteria, primarily Enterococcus spp., has been described (P. Linden, D. Parkinson, A. W. Pasculle, M. Birmingham, G. Zurenko, B. Hafkin, and D. Batts, Abstr. 39th Intersci. Conf. Antimicrob. Agents Chemother., abstr. 1105, 1999; T. K. Chowdhry, S. H. Marshall, C. J. Donskey, R. A. Salata, and L. B. Rice, Abstr. 100th Meet. Am. Soc. Microbiol. abstr. A63, 2000; S. M. Swaney, D. Shinabarger, R. Schaadt, J. Bock, J. Slightom, and G. Zurenko, Abstr. 38th Intersci. Conf. Antimicrob. Agents Chemother., abstr. C-104, 1998). These cases have defined point mutations involving the rRNA gene. This is a point of concern with strains of M. chelonae and M. abscessus, which have only a single copy of the ribosomal operon (1, 7, 22) and for which single point mutations in the ribosomal gene following monotherapy would likely result in drug resistance. Resistance from such mutational changes in the ribosomes of these two species has been described with clarithromycin (A2058 or A2059 in the 16S rRNA gene) (16, 22) and amikacin (A1408 in the 23S r-RNA gene) (13). Previous studies have shown that mutational ribosomal resistance with the macrolides is a problem only in the setting of disseminated cutaneous disease or pulmonary disease, but not with localized wound infections. Thus, linezolid should be used in combination therapy whenever possible with treatment of M. chelonae and M. abscessus, in the setting of disseminated or pulmonary disease. There is less concern with isolates of the M. fortuitum group, as they have two copies of the ribosomal operon (1, 7) and heterologous resistance with amikacin is recessive.

A serious limitation on therapy of disease due to these organisms is the cost of the drug, which is approximately $45 for a single 600-mg tablet, with a daily dosing schedule of 600 mg twice daily. This likely will mean that only patients in desperate clinical straits or in whom other drugs have failed will receive treatment, which is a difficult situation in which to determine clinical efficacy. At present no phase four clinical trials or compassionate trials of treatment of mycobacterial disease with linezolid are planned or are under way in the United States, although the drug was available on a compassionate basis prior to approval by the Food and Drug Administration. Other oxazolidinones are in trial or under consideration (R. C. Gadwood, L. M. Thomasco, E. A. Weaver, G. E. Zurenko, and C. W. Ford, Abstr. 39th Intersci. Conf. Antimicrob. Agents Chemother., abstr. 571, 1999). Hopefully these will come with a lower price tag, which will allow greater investigation of this very promising drug for disease caused by rapidly growing mycobacteria.