Background

Non-tuberculous mycobacteria (NTM) are mycobacteria other than M. tuberculosis and M. leprae. They are environmental pathogens predominantly found in water, soil, dust and animal sources [1]. In the past, NTM infections did not attract much attention since they did not exhibit human-to-human transmission. Very often, isolates were considered as contaminants by laboratories, and were therefore overlooked [2, 3].

Moreover, their epidemiology has been difficult to establish because reporting of NTM infections was not mandatory in most parts of the world [4]. Advanced diagnostic laboratory techniques, as well as increased clinicians’ awareness certainly contribute to a recent increase in worldwide reporting of their incidence, which is estimated to be between 1.0 and 1.8 per 100,000 people per year [5]. However, one determining factor behind the evolution of the epidemiology of NTMs is the emergence of the human immunodeficiency virus. In fact, people living with HIV, cystic fibrosis (CF), bronchiectasis, emphysema and chronic obstructive pulmonary disease (COPD) constitute the high-risk population for NTM infections [3, 5,6,7].

There are more than 170 NTM species identified to date, but only a few species are believed to be opportunistic pathogens and responsible for human infections. The most-common of these species are M. avium complex (MAC), M. abscessus complex (MABC), M. kansasii, M. fortuitum, M. chelonae, M. szulgai, M. triviale and M. scrofulaceum.

Infections caused by NTM are diverse and present in both immunocompromised and immunocompetent patients. Bone infections, skin infections and disseminated disease have been described; 90% of which are chronic lung infections in the same way as tuberculosis [8]. In fact, clinically, it is difficult to discern the symptoms of NTM lung disease and tuberculosis. This can lead to incorrect diagnoses in the absence of microbiological identification. Moreover, NTM are resistant to several antimycobacterial agents; hence the low efficacy of the latter [9, 10].

In order to inform clinical decision-making to optimise treatment efficacy, as the intrinsic antibiotic resistance profile is species-specific [5, 11], species identification must precede sensitivity profile determination. Macrolides and aminoglycosides have been considered first-choice molecules in the management of NTM infection according to the American Thoracic Society/Infectious Diseases Society of America (ATS/IDSA) [5].

In high-income countries, the management of NTM can be considered routine [5]. However, in the sub-Saharan region, there is very little data on NTM management and drug susceptibility profiles, as very few laboratories perform these tests. Most studies are focused on the prevalence of pulmonary NTM. In 2017, a systematic review and meta-analysis of 37 articles on NTM in the South-of the-Sahara region revealed a prevalence of 7.5% of pulmonary NTM, with MAC as the predominant species [12].

Like the other sub-Saharan countries, Gabon has some data on the isolation of NTM. Indeed, a study conducted by our group in 2022 showed a proportion of 10% of NTM in patients presumed to have tuberculosis in the Lambaréné region, with MAC being the most-common species [13]. Whilst antimicrobial susceptibility information is essential for clinicians to select appropriate treatment regimens, drug susceptibility testing for NTM species has not been widely conducted in Gabon.

Thus, the objective of this study was to identify the species and drug susceptibility profiles of NTM commonly isolated at the Tuberculosis Reference Laboratory of the Lambaréné Medical Research Center (CERMEL).

Methods

Study site

This study was carried out at the CERMEL TB laboratory, which is the national TB reference laboratory of Gabon. The laboratory receives samples from laboratories in peripheral areas for routine surveillance of drug-resistant tuberculosis. In addition, the laboratory performs susceptibility testing (Line Probe Assay and conventional DST) and monitors MDR-TB patients from the whole country using a culture technique [14,15,16].

Study design

This was a prospective cross-sectional study of isolates from positive cultures of presumptive TB patients identified as NTM species, from January 2020 to December 2022. A limited number of specimens collected in 2020 in the framework of an earlier-reported study were also included in this analysis [13].

Sampling

Two sputum samples of each presumptive TB patient were collected over two consecutive days. Specimens that were positive after culture were specified as NTM using Genotype CM/AS version 2.0 (Bruker Hain Lifescience, Nehren, Germany). Only specimen from patients with species found to be identical in each of the culture isolates were eligible for susceptibility testing.

Data collection

All data were collected from patients using a structured questionnaire and on the Laboratory test request form. Demographic data (age, sex), clinical information (HIV status, history of TB treatment, clinical symptoms), AFB smear results, and radiological results were collected. Study data were managed using REDCap (Research Electronic Data Capture) [17, 18].

Extraction and species identification

DNA was extracted from a positive mycobacterial culture using the Genolyse kit (Hain Lifescience, Nehren, Germany). The GenoType CM/AS kit (Hain Lifescience, Nehren, Germany) was used to prepare the master mix for polymerase chain reaction (PCR) amplification of the speciation-determining region using biotinylated primers present in the kit. After amplification, the labelled PCR products are hybridised with specific oligonucleotide probes immobilised on a strip. The post-hybridisation reaction results in the appearance of coloured bands on the band at the probe binding site and can be observed with the naked eye. Labelled hybrids captured and detected by colorimetric development can be used to detect different species. All the assays were performed according to the manufacturer's instructions [19].

Genotypic drug susceptibility testing of NTM (GenoType NTM-DR test)

In vitro antimicrobial susceptibility by the broth microdilution testing of NTM was not performed within the realm of this study. We performed the genotypic method with the GenoType NTM-DR assay, a qualitative in vitro test assay based on PCR and DNA strip technology. Mycobacterial DNA is extracted from cultivated material, specifically amplified via PCR, and detected on membrane strips using reverse hybridisation and an enzymatic colour reaction. The Genotype NTM-DR assay permits the simultaneous genetic detection of several relevant NTM: M. avium, M intracellulare, M chimaera, M. chelonae and the M. abscessus complex (M. abscessus subsp. abscessus, M. abscessus subsp. massilliense and M. abscessus subsp. bolletii); their resistance to aminoglycosides (kanamycin, amikacin, gentamicin) via the detection of the most relevant mutations of the rrs gene; and their resistance to macrolides (clarithromycin, azithromycin) via the detection of the most relevant mutations of the rrl gene. Additionally, the erm(41) gene is analysed for the identification of macrolide resistance in members of the M. abscessus complex. The gene is divided into two probes; the erm(41) C28 probe and the erm(41) T28 probe. When the erm(41) C28 probe is positive, this indicates that the strain tested is susceptible to macrolides (except for strains with the additional rrl mutation). When the erm(41) T28 probe is positive, this indicates that the strain tested is resistant to macrolides, and the rrl gene is also examined for the detection of resistance to macrolides (clarithromycin or azithromycin). The rrs gene is examined for polymorphisms indicative of resistance to aminoglycosides (kanamycin, amikacin, gentamicin).

Data analysis

Data analysis was performed using R version 4.0.1; the dplyr package was used for calculus methods. Chi-square tests were performed to establish a relationship between the categorical or binary variables. The factors associated with positive NTM status were quantified by logistic regression. The model building process followed a backward stepwise strategy whereby univariable analysis of all variables preceded the multivariable analysis. A variable was retained in the model if it was significant (p < 0.05). Two-sided p-values of 0.05 or less were considered statistically significant.

Study results

NTM species identification

The specimens in this study came from 524 culture positive samples among presumptive TB patients, including 29 fast-growing samples collected in 2020 from a previous study [13]. A total of 146 isolates (27.9%) were identified as NTM between February 2020 to November 2022. Of the 146 isolates confirmed as NTM, 142 (97.3%) were from sputum, while four (2.2%) were from urine samples. Four species (complexes) were identified among the isolates, and the predominant group was M. avium complex (MAC 80/146; 54.8%; of which M. intracellulare 53/146 (36.3%) and M. avium 27/146 (18. 5%)). M. abscessus complex (MABC) amounted to 38/146 (26.0%) of the isolates. However, in 11 cases (7.5%), there were two species detected. GenoType Mycobacterium NTM-DR could identify the subspecies of MABC M. abscessus subsp. abscessus (20 isolates), M. abscessus subsp. bollettii (8), and M. abscessus subsp. massiliense (10). The isolated species are listed in Table 1.

Table 1 Distribution of banding pattern and NTM species among 146 presumptive TB patients analysed
Full size table

Patient characteristics

The average age of study participants was 46.8 (SD:14) years; 88/142 (62.0%) of the patients were male; 47/142 (33.1%) patients were HIV-co-infected. Regarding treatment history, 32/142 (22. 5%) of the NTM patients isolates had previously been treated for TB. Among the positive NTM isolates, only 34/142 (23.9%) of the smears were positive (Table 2). The clinical characteristics of NTM infections of male and female patients are compared in Table 3. The results show that the proportion of cough and hemoptysis in male patients were significantly higher than in female patients (93% versus 36%, p < 0.030; 37% vs 20%, p < 0.040, respectively).

Table 2 Characteristics of the 142 patients found positive to the NTM among presumptive TB patients
Full size table
Table 3 Comparison between male and female pulmonary NTM isolates patients among presumptive tuberculosis patients
Full size table

Genotypic drug susceptibility testing of NTM species

Of the 146 NTM isolates, 135 were tested for antimicrobial susceptibility to macrolides (azithromycin, clarithromycin and capreomycin) and aminoglycosides (kanamycin, amikacin, and gentamycin) using GenoType NTM-DR. The other 11 species were not tested due to the multiple NTM species identified in the same isolate. Of the 38 isolates of the subspecies of the MABC complex, five isolates of the subspecies M. abscessus subsp. abscessus were found to be resistant to macrolides, with a mutation present in T28 in the erm probe (41), and a mutation in the rrl gene. Resistance to aminoglycosides was also found by a mutation in the rrs gene. The five strains that presented the erm(41) mutation also carried rrl and rrs mutations. However, no resistance-conferring mutations were detected for either MAC complex or M. chelonae (Table 4).

Table 4 Associated mutations and drug susceptibility results for macrolides and aminoglycosides in the GenoType1 Mycobacterium NTM-DR assay
Full size table

Discussion

This study identified NTM species in presumptive tuberculosis patients in Gabon and determined their drug susceptibility profile using the LPA GenoType Mycobacterium NTM-DR test. The study revealed that M. intracellulare was the most-commonly occurring species, as observed previously in a study done in the same region, and also demonstrated in a study involving several countries in Sub-Saharan Africa [12, 13]. However, the MABC was the second most-common species isolated, which is contrary to the previous study where M. fortuitum was the second most-frequently found species. This describes the diversity of NTM species circulating among patients presumed to have tuberculosis in the Lambaréné region. As well, it has been documented that 25 to 60% of patients with a positive respiratory specimen for microbiological, respiratory signs and radiographic criteria have NTM pulmonary disease, and that due to its variable virulence, it is important to identify the species of NTM for better patient management [20].

In this study, results show that NTM isolates were more prevalent in male patients (88/142; 62%) as compared to female patients. This is in line with studies performed in Europe with 70%, and Tanzania 66%, which have suggested NTM pulmonary isolates to be originating mostly from men [21, 22]. However, some studies have indicated that pulmonary disease from NTM affects more women than men [23,24,25]. The findings from our study can be explained by the fact that our study population is based on cases of presumptive tuberculosis patients, and that in most studies on tuberculosis the male gender is the most-affected by the disease [22, 26]. Also, having a history of tuberculosis for pulmonary tuberculosis has been reported to be associated with NTM pulmonary isolates [13]. In this study, about 25% of the patients with NTM pulmonary isolates had a history of confirmed tuberculosis. Radiologically, the pulmonary presentation of NTM in our study was characterized by the presence of infiltrations and cavitations, which are also radiographic features similar to those of tuberculosis [27].

According to the Clinical Laboratory Standards Institute (CLSI) recommendations, drug susceptibility testing of NTM is usually done by the broth microdilution method as the gold standard. However, this method is time-consuming, given the delay due to the slow mycobacterial growth, as it is the case in M. avium [11]. In our study, 5/38 isolates of the MABC group were resistant to macrolides, with a mutation in the erm(41) gene at position T28 and mutation in the rrl gene in M. abscessus subsp. abscessus. These findings are similar to those of Maya et al., which is due to the fact that M. abscessus subsp. abscessus and M. abscessus subsp. bolettii macrolide resistance is identified by polymorphisms at position 28 in the erm(41) gene (cytosine is replaced by thymine leading to resistance) and mutations at positions 2058/2059 (adenine is replaced by cytosine leading to resistance) in the rrl gene [22, 28, 29]. However, other species of this group such as M. massiliense have a non-functional erm(41) gene, do not have the target base at position 28 which results from a large deletion of a non-functional 276 bp, and therefore do not exhibit resistance to macrolides. To determine NTM macrolide resistance, it is necessary to check whether there is a mutation in the rrl gene [30, 31]. Five isolates of M. abscessus subsp. abscessus were resistant to aminoglycosides through the rrs gene. Management of MABC group NTM that are resistant to macrolides and aminoglycosides should be performed according to guidelines [32]. In our study, no mutation was detected in the MAC group (M. avium, M. intracellulare) indicating that MAC is sensitive to macrolides and aminoglycosides; in line with a study conducted in Ghana showing sensitivity of all MAC isolates to macrolides and aminoglycosides drugs [33].

This study has limitations. Firstly, the presence of two species of NTM in an isolate does not allow the genotypic susceptibility test to be carried out for those type of isolates, and the genotype can only be used for the species indicated in the kit, while the reference method (microdilution) can be used as a complement for the other species. Secondly, although macrolides and aminoglycosides are the first-line molecules for NTM lung disease, other molecules that can be combined with the therapeutic regimens have not been taken into account by the test used.

Conclusions

Our study identified NTM species and tested drug susceptibility in routine culture isolates among tuberculosis patients in Lambaréné, Gabon. The study confirmed that the predominant NTM species was M. intracellulare, and that all MAC species were sensitive to macrolides and aminoglycosides. On the other hand, five MABC group isolates linked to the subspecies M. abscessus subsp. abscessus showed genotypical resistance to macrolides and aminoglycosides. Our research shows that NTM are present among presumptive TB cases in our setting, hence National Tuberculosis Control Programs should consider screening for NTM among presumptive TB cases. Knowledge of the NTM infection status and NTM drug susceptibility pattern is essential for correct diagnosis and selection of appropriate treatment regimens and should guide clinical decision-making.