Evaluation of the GenoType NTM-DR assay performance for the identification and molecular detection of antibiotic resistance in Mycobacterium abscessus complex

The first objective of this study was to determine the GenoType NTM-DR assay performance for subspecies identification in Mycobacterium abscessus complex isolates. The second objective was to evaluate the GenoType NTM-DR assay ability to detect clarithromycin and amikacin resistance in M. abscessus complex isolates compared with drug susceptibility testing (DST) and PCR sequencing of the erm(41), rrl and rrs genes. The concordance between the GenoType NTM-DR and MLST results concerning subspecies identification was 100%. The wild type and mutated alleles of the rrl and rrs genes were detected by the GenoType NTM-DR assay and PCR sequencing with 100% (115/115) agreement. Similarly, 100% concordance between GenoType NTM-DR and DST was observed for clarithromycin and amikacin testing. Sensitivity for the detection of clarithromycin and amikacin resistance was 100%. The GenoType NTM-DR assay provides a robust and complementary tool to the gold standard methods (MLST and broth microdilution) for subspecies identification and drug resistance detection.


Introduction
The fast-growing Mycobacterium abscessus complex is an emerging opportunistic human pathogen worldwide [1]. It includes three subspecies: M. abscessus subspecies abscessus, M. abscessus subsp. massiliense, and M. abscessus subsp. bolletii [2]. The most frequent clinical manifestation of M. abscessus complex infection is chronic lung infection, usually in elderly women with bronchiectasis and in young adults with cystic fibrosis (CF) [3].
M. abscessus carries or acquires inducible resistance to macrolides (e.g. clarithromycin) and aminoglycosides (e.g. amikacin) used to treat pulmonary infections. Resistance to macrolides can be: i) constitutive (associated with point mutations at position A2058 and A2059 of the rrl gene, a macrolide target), in all subspecies; and ii) inducible (associated with the erm(41) gene T28 sequevar), only in M. abscessus subsp. abscessus and bolletii [4]. A single 16S ribosomal RNA (rRNA) substitution at position 1408 in the rrs gene causes high-level aminoglycoside resistance [4,5]. Rapid molecular diagnosis is essential for subspecies identification and determination of the adequate antimicrobial therapy, and is now recommended in patients with CF [3,6].
This study wanted to 1) determine NTM-DR performance for subspecies identification in M. abscessus complex isolates (compared with MLST), and 2) evaluate NTM-DR ability to detect clarithromycin and amikacin resistance (compared with phenotypic and sequencing analyses).

Patients and mycobacteria isolates
For this study, 115 M. abscessus complex isolates (1 isolate/patient) were randomly selected among the 176 isolates (1 isolate/patient) from respiratory samples of 176 different patients with CF or other respiratory diseases stored at the Microbiology Laboratory, University Hospital, Montpellier (France). All M. abscessus complex isolates came from six University Hospitals in France (Brest, Bordeaux, Montpellier, Nantes, Rouen, and Toulouse), including a Reference Centre for CF.
Resistance was assessed using the GenoType NTM-DR Kit and by phenotypic drug susceptibility testing (DST), the reference method [11]. The Minimum Inhibitory Concentrations (MICs) of clarithromycin and amikacin were evaluated using the reference broth microdilution method with Sensititre RAPMYCO microplates (Trek Diagnosis Systems). Susceptibility and resistance were assessed according to the CLSI recommendations. High-level clarithromycin resistance was defined by a MIC �8 μg/ml at day 5. Inducible resistance was defined by an increase in clarithromycin MIC from �2 μg/ml at day 5 to �8 μg/ml at day 14. Aminoglycoside resistance was defined by MIC values �64 μg/ml [11]. Mutations in the erm (41), rrl and rrs genes were identified by PCR and sequencing, as previously described [12,13].
NTM-DR detects heteroresistance in mixed populations of drug-susceptible and -resistant mycobacteria and shows the simultaneously the presence of wild type (WT) and mutant (MUT) bands.
NTM-DR identified rrl mutations associated with high level of resistance in the 12 isolates with high resistance to clarithromycin (n = 4 A2058C, n = 5 A2058G, and n = 3 A2059G rrl mutation, corresponding to the positive MUT1, MTU2 and MUT4 bands of NTM-DR, respectively, (S1 Table). Among the 12 clarithromycin-resistant M. abscessus subsp. abscessus isolates, 11 showed a rrl profile without WT band and one presented a heterogeneous pattern (one WT and one MUT1 band), suggesting the presence of a mixed population. In all 47 isolates with inducible clarithromycin resistance, NTM-DR detected a specific erm (41) T28 sequevar band and a WT band in the rrl gene. Finally, all 56 isolates susceptible to clarithromycin by DST were susceptible also by NTM-DR (erm(41) C28 sequevar band and WT band in the rrl gene) (S1 Table).
The six isolates with aminoglycoside resistance displayed the positive MUT1 band by NTM-DR. Sequencing confirmed the presence of the A1408G rrs mutation in these isolates. The 109 (94.8%) aminoglycoside-susceptible isolates showed a WT band by NTM-DR and WT rrs gene sequence (S1 Table). Overall, NTM-DR and sequencing showed 100% agreement. The concordance between NTM-DR and DST was 100% for clarithromycin and amikacin. NTM-DR sensitivity for the detection of clarithromycin and amikacin resistance was also 100%.

Discussion
Our study showed an excellent concordance (100%) between NTM-DR and MLST results concerning subspecies identification, and confirmed the high sensitivity of NTM-DR for detecting acquired resistance. This is higher than what reported by a recent study on 50 M. abscessus complex isolates (92% concordance between NTM-DR and gene sequencing for subspecies identification) [7,14,15]. The excellent results observed confirmed that the NTM-DR assay seems a discriminative method for M. abscessus complex subspecies identification [7,9,16].
Our study showed that NTM-DR can detect acquired resistance with high sensitivity. This test also easily detects heteroresistance, not always possible by sequencing [8]. This is crucial for treatment decision-making, and it would be relevant to evaluate in vitro NTM-DR sensitivity for heteroresistance detection in mixed populations of drug-susceptible and -resistant mycobacteria.
Although this is not the case for our sample, it has been reported that some isolates with phenotypic resistance to macrolides and aminoglycoside do not carry mutations in the rrl and rrs genes. This implies the existence of other resistance mechanisms that cannot be detected by the NTM-DR and other commercial kits [7,8]. For example, a modification (direct repeat 18-bp insertion in rpIV) in the ribosomal protein L4 has been recently associated with resistance to macrolides [17].

Conclusion
The excellent NTM-DR performance indicates that this test is a robust and complementary tool to MLST and broth microdilution for subspecies identification and detection of clarithromycin and amikacin resistance in M. abscessus complex. NTM-DR could be used for routine testing, particularly in patients with and without CF and M. abscessus complex lung infections.