Overall, only 28 isolates were identified as M. species or mixed NTM by GenoType Mycobacterium CM/AS and GenoType NTM-DR (all three tests performed for each isolate) during the 5-year study period in the Czech Republic and Slovakia. The occurrence of NTM species that could not be identified using GenoType was thus relatively low in both Slovakia and the Czech Republic. Slovakia recorded 9 such cases in 2023, while the Czech Republic reported 1 case in 2019, 4 in 2020, 6 in 2021, 1 in 2022, and 7 in 2023. In the Czech Republic, the proportion of this NTM species among all identified NTM (except M. gordonae) was 0.9% (1/106) in 2019, 3.6% (4/112) in 2020, 6.5% (6/92) in 2021, 1.1% (1/95) in 2022, and 3.8% (7/185) in 2023. In Slovakia, 7 isolates of M. species represent 4.4% of the total number of identified NTM (except M. gordonae; n = 158). Among the patients, 15 (53.6%) were females, and 13 (46.4%) were male. The median age of the patients was 65.4 ± 20.7 years (ranged 2 to 87). Out of 28 isolates, 25 (88.9%) were obtained from sputum, 2 (7.4%) from lymph node biopsy and 1 (3.7%) from urine. Most patients were diagnosed with lung diseases (Fig. 1, Supplementary Table 1).

A summary of the initial diagnostic assessment upon which sample collection was based

The data obtained by Deeplex Myc-TB allowed species classification of 27 out of 28 (96.4%) isolates. In total, 13 different species were reported, with M. neoaurum the most prevalent species (n = 8), followed by M. kumamotonense (n = 5), M. arupense (n = 5) and M. chimaera (n = 3). In one isolate, no mycobacteria were detected (Fig. 2). Six samples contained mixed NTM species, two of which were also positive for multiple NTM species by GenoType NTM-DR (Supplementary Table 1). The other sample initially identified as mixed NTM species by GenoType NTM-DR was characterized solely as M. chimaera using Deeplex Myc-TB.

NTM species identified by Deeplex Myc-TB

Due to insufficient DNA concentration and the inability to acquire additional culture material for further DNA extraction, WGS was only performed on 19 isolates. Among the 9 isolates without WGS data, 5 were identified as M. neoaurum, 2 as M. arupense, 1 as a mixed NTM species (M. arupense/M. yongonense), and 1 as M. holsaticum by Deeplex Myc-TB. A comparison of Deeplex Myc-TB and WGS results revealed discrepancies in 7 out of 19 isolates (Table 1). Four of these isolates were initially identified as mixed NTM species by Deeplex Myc-TB. However, WGS analysis did either not detect any NTM but instead Tsukamurella tyrosinosolvens (n = 1) or detect only 1 NTM species (n = 3) in these samples. Based on WGS, potential novel NTM species were found in 3 samples. However, Deeplex analysis identified these as mixed samples or as M. arupense. In the sample where no mycobacteria were detected with Deeplex Myc-TB, WGS identified M. celatum. In addition, WGS reported M. branderi in one sample, whereas Deeplex Myc-TB reported a mix of M. chimaera and M. massiliense (i.e. M. abscessus subsp. massiliense). Based on the above, the agreement rate of WGS and Deeplex Myc-TB was 63.2%. The agreement rate between TYGS and NTMprofiler was 100% for the 12 isolates where TYGS reported an NTM.

To further classify the potentially new species, a GBDP tree was calculated, incorporating all closely related Mycobacterium reference genomes (Fig. 3).

Tree inferred with FastME 2.1.6.1 from GBDP distances calculated from genome sequences. The branch lengths are scaled in terms of GBDP distance formula d5. The numbers above branches are GBDP pseudobootstrap support values > 60% from 100 replications, with an average branch support of 81.5%. The tree was rooted at the midpoint. Red crosses indicate novel NTM species. Three highly contaminated samples (CZ11223, CZ61421 and CZ45720) were not included. User strain refers to the isolates obtained in this study. The blue plus (+) symbol represents previously identified NTM species, while the red plus (+) symbol indicates novel NTM species

Genome-to-genome distances were computed based on dDDH and ANI to assess species-level boundaries. Based on dDDH formula d0 and d4, the closest reference genome to strain CZ13121 is M. wolinskyi, a rapidly growing mycobacterium (RGM). The ANI compared to the M. wolinskyi type strain ATCC700010 was only 86.3% with 75.0% alignment fraction (AF), indicating that CZ13121 represents a distinct species within the Mycobacterium genus.

For strain CZ26520, dDDH formula d0 suggests that the closest reference genome is M. senuense, with M. heraklionense as the second closest and not M. arupense as suggested by Deeplex Myc-TB. In addition, using the dDDH formula d4, which is considered more reliable for draft genomes, the closest related genome is M. heraklionense, a slow-growing mycobacterium (SGM) belonging to the M. terrae complex. Similarly, for strain CZ46821, both dDDH formulas d0 and d4 indicate that the closest reference genome is M. heraklionense but ANI with this strain is still < 95%. Moreover, as CZ26520 and CZ46821 show a pairwise ANI of 98.6% and AF of 94%, the two strains can be considered as belonging to the same novel species.

Based on experimental growth rate, CZ26520 and CZ46821 were classified as slow-growing NTM, while CZ13121 was categorized as a fast-growing NTM.

Most isolates (22/28) were susceptible to amikacin, ciprofloxacin, moxifloxacin, clarithromycin and linezolid (Table 2). The most prevalent resistance was observed to ciprofloxacin (4/28). Among the resistant isolates, the novel species CZ131/21 demonstrated exceptional inducible resistance to clarithromycin, with a MIC exceeding 32 µg/ml, which might be attributed to the presence of a methyltransferase (erm) gene as detected by AMRfinder + with relaxed thresholds (Supplementary Table 2). The isolate M. branderi (CZ614/21) demonstrated resistance to both linezolid and ciprofloxacin. This patient met the ATS/ERS/ESCMID/IDSA criteria and, despite resistance, administering the combination of ethambutol, ciprofloxacin, and clarithromycin led to successful sputum culture conversion (Tables 2 and 3). M. kumamotonense (CZ126/21) from a patient with confirmed pulmonary mycobacteriosis exhibited resistance to moxifloxacin and ciprofloxacin, and intermediate susceptibility to linezolid, but treatment and outcome data were not available for this patient (Tables 2 and 3).

Five additional patients met the ATS/ERS/ESCMID/IDSA criteria for NTM pulmonary disease, and two cases involved pediatric patients presenting with typical cervical lymphadenitis (extrapulmonary NTM infection) caused by M. lentiflavum and M. celatum, yielding a substantial rate of confirmed mycobacterial infections (9 out of 27 patients; 33.3%) (Table 3). All four patients who fulfilled ATS/ERS/ESCMID/IDSA criteria and received a combination therapy of rifampicin, ethambutol, and clarithromycin showed culture conversion and, thus, positive treatment outcomes. In addition, one patient with lymphadenitis was successfully treated by surgical removal of the lymph node. At the same time, first-line anti-tuberculosis drugs with clarithromycin were effective in treating mixed infections with M. tuberculosis and M. avium (Supplementary Table 1). Furthermore, no relapse was observed in patients who achieved a successful treatment outcome as of December 2023.