Antimicrobial Susceptibility in the Mycobacteroides abscessus Complex is Restored by an Imipenem-Clarithromycin Combination

Nontuberculous mycobacteria (NTM) are ubiquitous organisms and the incidence of NTM infections has been increasing in recent years. Mycobacteroides abscessus (M. abscessus) is one of the most antimicrobial-resistant NTM; however, no reliable antibiotic regimen can be ocially advocated. We evaluated the ecacy of clarithromycin in combination with various antimicrobial agents against the M. abscessus complex. Twenty-nine clinical strains of M. abscessus were isolated from various clinical samples. Of the isolates, 10 (34.5%) were of M. abscessus subsp. abscessus, 18 (62.1%) of M. abscessus subsp. massiliense, and 1 (3.4%) of M. abscessus subsp. bolletii. MICs of three antimicrobial agents (amikacin, imipenem, and moxioxacin) were measured with or without clarithromycin. The imipenem-clarithromycin combination signicantly reduced MICs compared to clarithromycin and imipenem monotherapies, including against resistant strains. The association between susceptibility of the M. abscessus complex and each combination of agents was signicant (p = 0.001). Adjusted residuals indicated that the imipenem-clarithromycin combination had the synergistic effect (adjusted residual = 3.1) and suppressed the antagonistic effect (adjusted residual = -3.1). In subspecies of M. abscessus complex, the association with susceptibility of M. abscessus subsp. massiliense was similarly statistically signicant (p = 0.036: adjusted residuals of synergistic and antagonistic effect respectively: 2.6 and -2.6). The association with susceptibility of M. abscessus subsp. abscessus also showed a similar trend but did not reach statistical signicance. Our data suggest that the imipenem-clarithromycin combination could be the recommended therapeutic choice for the treatment of M. abscessus complex owing to its ability to restore antimicrobial


Abstract
Nontuberculous mycobacteria (NTM) are ubiquitous organisms and the incidence of NTM infections has been increasing in recent years. Mycobacteroides abscessus (M. abscessus) is one of the most antimicrobial-resistant NTM; however, no reliable antibiotic regimen can be o cially advocated. We evaluated the e cacy of clarithromycin in combination with various antimicrobial agents against the M. abscessus complex. Twenty-nine clinical strains of M. abscessus were isolated from various clinical samples. Of the isolates, 10 (34.5%) were of M. abscessus subsp. abscessus, 18 (62.1%) of M. abscessus subsp. massiliense, and 1 (3.4%) of M. abscessus subsp. bolletii. MICs of three antimicrobial agents (amikacin, imipenem, and moxi oxacin) were measured with or without clarithromycin. The imipenem-clarithromycin combination signi cantly reduced MICs compared to clarithromycin and imipenem monotherapies, including against resistant strains. The association between susceptibility of the M. abscessus complex and each combination of agents was signi cant (p = 0.001). Adjusted residuals indicated that the imipenem-clarithromycin combination had the synergistic effect (adjusted residual = 3.1) and suppressed the antagonistic effect (adjusted residual = -3.1). In subspecies of M. abscessus complex, the association with susceptibility of M. abscessus subsp. massiliense was similarly statistically signi cant (p = 0.036: adjusted residuals of synergistic and antagonistic effect respectively: 2.6 and -2.6). The association with susceptibility of M. abscessus subsp. abscessus also showed a similar trend but did not reach statistical signi cance. Our data suggest that the imipenem-clarithromycin combination could be the recommended therapeutic choice for the treatment of M. abscessus complex owing to its ability to restore antimicrobial susceptibility.
Background NTM are ubiquitous organisms that cause diverse types of infectious diseases in humans, including in lungs, the lymphatic system, skin, soft tissue, bone disease, and are disseminated. The morbidity of NTM has been increasing worldwide (1,2); the 2014 nationwide survey of NTM in Japan revealed that the incidence of pulmonary NTM (14.7 cases/100,000 person/year) has overtaken that of tuberculosis (12.9 cases/100,000 person/year) (3). Above all, the Mycobacterium avium complex (88.8%) were the most frequently isolated organisms, followed by Mycobacterium kansasii (4.3%) and the Mycobacteroides abscessus complex (3.3%). Notably, the incidence of M. abscessus-infected pulmonary disease has dramatically increased in Japan, from 0.1 cases/100,000 person/year in 2001 to 0.5 cases/100,000 person/year in 2014. M. abscessus is one of the treatment-refractory NTM, characterized by rapid growth and multidrug resistance. It is also frequently isolated from the respiratory tract of patients with cystic brosis (CF); M. abscessus has been the leading cause of rapid growing mycobacteria in CF since the 2000s (4,5). The critical feature of M. abscessus is its spontaneous resistance to most antibiotics in clinical use, including rst line antitubercular drugs (6,7). The 2007 American Thoracic Society/Infectious Diseases Society of America (ATS/IDSA) statement recommended multidrug therapy, including a macrolide and one or more parenteral agents (e.g., amikacin, cefoxitin, or imipenem) (

Susceptibility to antimicrobial agents in combination with clarithromycin
The susceptibility to a combination of clarithromycin and antimicrobial agents was compared to that of the antimicrobial agents alone, categorized into each subspecies of M. abscessus complex ( Figure 1). The MICs of three antimicrobial agents (amikacin, imipenem, and moxi oxacin) were measured with or without clarithromycin. Susceptibility to the imipenem-clarithromycin combination was signi cantly better than to other clarithromycin combinations. Notably, the use of the imipenem-clarithromycin combination signi cantly reduced the MIC of clarithromycin, even in clarithromycin-resistant subspecies of M. abscessus complex. The effect of restoring susceptibility by the imipenem-clarithromycin combination was stronger than that of the amikacin-and moxi oxacin-clarithromycin combination. In 1 strain of M. abscessus subsp. abscessus and 3 strains of M. abscessus subsp. massiliense, susceptibility was not restored by the combined use of clarithromycin and imipenem, and only 1 strain of M. abscessus subsp. bolletii did not respond to any combination. In subspecies of M. abscessus complex, MICs of imipenem and clarithromycin in combination were signi cantly less than that of either clarithromycin or imipenem alone in both M. abscessus subsp. massiliense and abscessus (p < 0.001 for both subspecies) (Table S1).
We next determined the synergistic effect of the imipenem-clarithromycin combination as compared to amikacin-or moxi oxacin-clarithromycin combinations, using the fractional inhibitory concentration (FIC) index ( Figure 2). Susceptibility was divided into two classes, synergy and additive as a synergistic effect and indifference and antagonism as an antagonistic effect. The associations between susceptibility of the M. abscessus complex and each combination of antimicrobials were signi cant (p = 0.001) ( Table 2). Adjusted residuals indicated that the imipenem-clarithromycin combination had the synergistic effect (adjusted residual = 3.1) and suppressed the antagonistic effect (adjusted residual = -3.1). In the subspecies of M. abscessus complex, the association between susceptibility of M. abscessus subsp. massiliense and each combination of antimicrobials was signi cant (p = 0.036), and imipenemclarithromycin combination had the synergistic effect (adjusted residual = 2.6) and suppressed the antagonistic effect (adjusted residual = -2.6). The association with susceptibility of M. abscessus subsp. abscessus also showed a similar trend, but did not reach statistical signi cance, potentially because of a smaller number of samples.

Association of clinical features with susceptibilities to the imipenem-clarithromycin combination
We investigated whether susceptibility to the imipenem-clarithromycin combination might associate with clinical status. The isolates from patients with immunosuppression and/or administered immunosuppressive drugs and/or corticosteroids revealed synergistic effects rather than antagonistic effects (p = 0.040) ( Table 3). The other clinical parameters such as age, sex, smoking history, bronchiectasis lesion, and a treatment history of antibiotics did not associate with the susceptibility to the imipenem-clarithromycin combination.

Discussion
We demonstrated here that the MICs of clarithromycin and imipenem were signi cantly reduced by the administration of an imipenem-clarithromycin combination. We propose a new therapeutic bene t by which the imipenem-clarithromycin combination could restore the susceptibility of M. abscessus isolates, even after acquiring resistance to both clarithromycin and imipenem separately. The isolates included M. abscessus subsp. abscessus, well known among the three subspecies to easily acquire macrolideresistance (9,10). Furthermore, this combination may be suitable for treatment of M. abscessus complex in patients with immunosuppression.
In recent years, the incidence of NTM has globally increased. The major issue in treating M. abscessus complex centers on its intrinsic resistance against most available antibiotics. The 2007 ATS/IDSA Statement has recommended antimicrobial combination therapy, namely macrolides (clarithromycin), aminoglycosides (amikacin), cephamycins (cefoxitin), and carbapenems (imipenem), to treat M. abscessus infections, depending on in vitro susceptibility testing (8). However, there were several problems involved in the recommendation, due to the lack of clinical outcomes, and uncertain interactions present in multidrug combination therapy; thus, there is still limited reliable evidence to promote a global standard treatment regimen for the three subspecies of M. abscessus complex. Previous in vitro studies have demonstrated that treatment with the standard regimen therapy (combinations of clarithromycin, amikacin, and cefoxitin) failed to effectively inhibit the growth of M. abscessus due to acquired drug resistance (16). In vivo, the triple-drug regimen was equally or less effective against M. abscessus than cefoxitin alone (17). A systematic review revealed different outcomes of macrolide-containing combination regimens against M. abscessus subsp. abscessus and massiliense. Macrolide-containing combination regimens for M. abscessus subsp. abscessus induced lower rates of negative conversion of sputum culture and higher recurrence rates than that of M. abscessus subsp. massiliense (18). For these reasons, the appropriate drug therapy against M. abscessus remains uncertain. M. abscessus complex spontaneously produce broad-spectrum βlactamases, resulting in reduced susceptibility to β-lactams, including imipenem. Imipenem in combination with rifabutin or amikacin was more effective than as a monotherapy against M. abscessus complex (19,20). Miyasaka et al. described that the imipenem-clarithromycin combination had a high rate of synergistic and additive effects, and revealed a decrease in the MIC values inhibiting 50% or 90% of M. abscessus complex (21). Although the exact mechanism for the synergistic effect of clarithromycin combinations was unknown, these data support the results of our present study. Therefore, imipenem may be useful in combination with clarithromycin for the treatment of M. abscessus complex. Limitations of the present study include the lack of clinical outcomes measured in patients with M. abscessus complex treated with imipenem-clarithromycin combination therapy. Thus, a prospective clinical study is required to establish the in vivo e cacy of the combination regimen.  (17,18). All strains of M. abscessus were cultured on BD trypticase soy agar II with 5% sheep blood (Blood agar; Nippon Becton-Dickinson and Company, Japan) at 35°C for approximately 4 to 6 days in an aerobic atmosphere. The study protocol was approved by the Ethics Committee of Juntendo University School of Medicine (no. 18-010 and 19-038).

MALDI-TOF MS analysis
Colonies of M. abscessus complex on blood agar were scratched with a needle, and particles on the needle surface were diluted in 50 μL 80% tri uoroacetic acid. After incubation for 15 minutes at room temperature, the solution was added to 150 μL distilled water and 200 μL 100% acetonitrile, followed by a centrifugation step (16,200 × g, 2 min). One microliter of the cleared supernatant containing the bacterial extract was transferred onto a MALDI target plate (Bruker Daltonik, Germany). Dried spots were overlaid with MALDI matrix (10 mg/mL α-cyano-4-hydroxy-cinnamic acid [α-HCCA] in 50% acetonitrile:2.5% tri uoroacetic acid) (Bruker Daltonik, Germany). After drying of the matrix, MALDI-TOF MS measurements were performed with a Micro ex LT/SH benchtop mass spectrometer (Bruker Daltonik, Germany) equipped with a 60-Hz nitrogen laser. Parameter settings (ion source 1 [IS1], 20 kV; IS2, 18.2 kV; lens, 6.85 kV; detector gain, 2854 V; gating, none) had been optimized for the mass range between 2,000 and 20,000 Da. Spectra were recorded in the positive linear mode with the maximum laser frequency. An external standard (bacterial test standard [BTS]) (Bruker Daltonik, Germany) was used for instrument calibration. Data evaluation was performed by visually comparing spectra to search for peak shifts using exAnalysis 3.4 (Bruker Daltonik, Germany).
PCR ampli cation and DNA sequencing DNA was extracted from cultured colonies using the DNeasy UltraClean Microbial Kit (QIAGEN, Germany), and PCR was conducted using Ex Taq DNA polymerase, hot-start version (Takara, Japan) according to the manufacturer's instructions. The gene-speci c primer pairs used for PCR analysis are listed in Table 4; these primers were used in previous studies (19,20). The sequencing PCR products were puri ed with the BigDye XTerminator puri cation kit (Life Technologies, USA) and samples were loaded on the ABI Prism 3130 Genetic Analyzer (Thermo Fisher Scienti c, USA). The DNA sequencing results were analyzed using a BLAST search to identify sequence similarity between samples and the three species of M. abscessus complex.
Antimicrobial susceptibility testing Susceptibility testing was performed according to Clinical and Laboratory Standard Institute (CLSI) guideline M24-A2 (21), using frozen broth microdilution panels. The ranges of antibiotic concentrations tested were as follows: amikacin (AMK) 0.25 to 64 μg/mL, clarithromycin (CLR) 0.06 to 64 μg/mL, imipenem (IPM) 4 to 32 μg/mL, and moxi oxacin (MXF) 1 to 32 μg/mL. MICs of each antimicrobial agent were determined by broth microdilution methods, as recommended by the CLSI, using frozen broth microdilution plates (Eiken Chemical Co., Ltd., Japan). The MICs were determined after 7 days of incubation at 35°C. The MIC breakpoints, indicating susceptible, intermediate, and resistant strains, were interpreted according to the CLSI criteria for amikacin, cefoxitin, cipro oxacin, clarithromycin, doxycycline, imipenem, linezolid, moxi oxacin, trimethoprim/sulfamethoxazole, and tobramycin (Table 5) (21). The effect of each agent combined with clarithromycin was evaluated using FIC index analysis. FIC index was calculated as follows: Σ (FIC A + FIC B), where FIC A is the MIC of compound A in combination / MIC of compound A alone, and FIC B is the MIC of compound B in combination / MIC of compound B alone. The combination is considered synergistic when the Σ FIC is ≤0.5, additive when the Σ FIC is >0.5 to ≤1, indifferent when the Σ FIC is >1 to ≤2, and antagonistic when the Σ FIC is >2.

Statistical analysis
Categorical variables were compared using the chi-square test or Fisher's exact test. The evaluation of changes in MIC was performed using the Wilcoxon signed-rank test. Differences were considered signi cant at p <0.05. When the chi-square test results were statistically signi cant, adjusted residuals were calculated to determine which particular associations were signi cant. Adjusted residuals were signi cant at p < 0.05 level if they were less than −1.96 or more than 1.96, and were signi cant at p < 0.01 level if they were less than −2.58 or more than 2.58. All statistical analyses were performed using the SPSS software program (version 20, IBM Japan, Japan). The study was approved by the Independent Ethics Committee at Juntendo University Hospital (approval no. 18-010 and 19-038) and adhered to the tenets of the Declaration of Helsinki.

Consent for publication
Not applicable.

Availability of data and materials
The datasets during the current study available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. Supplementarytable20200127.pptx