ABSTRACT
Accurate species identification is a prerequisite for successful management of tuberculosis and non-tuberculous mycobacterial (NTM) diseases. The novel FluoroType Mycobacteria assay combines three established GenoType DNA strip assays (CM, AS, and NTM-DR), allowing detection of Mycobacterium tuberculosis and 32 NTM species/subspecies in a single assay with automatic detection and result analysis. We evaluated the clinical performance of the FluoroType assay and its feasibility in replacing the GenoType Mycobacterium CM assay as the initial method for mycobacterial identification. A total of 191 clinical mycobacterial cultures were analyzed in this study: 180 identified for one mycobacterial species, 6 for multiple, and 5 for no mycobacterial species. Positive percent agreement (PPA) for the FluoroType assay was 87.8% (n = 158), with full agreement for 23/29 species. Weakest PPA was observed for Mycobacterium gordonae (50%, n = 9/18), Mycobacterium interjectum (40%, n = 2/5), and Mycobacterium intracellulare (42%, n = 5/12). Clinical and mixed cultures containing multiple mycobacterial species gave equally single species and genus level identifications (n = 30). No cross-reactivity with non-mycobacterial species was observed (n = 22). In a separate in silico analysis of 2016–2022 HUS area (Finland) register data (n = 2,573), the FluoroType assay was estimated to produce 18.8% (n = 471) inadequate identifications (genus/false species) if used as the primary identification method compared to 14.2% (n = 366) with the GenoType CM assay. The FluoroType assay was significantly more convenient in terms of assay workflow and result interpretation compared to the entirely manual and subjective GenoType CM assay. However, the feasibility of the assay should be critically assessed with respect to the local NTM species distribution.
IMPORTANCE
This study is the first clinical evaluation report of the novel FluoroType Mycobacteria assay. The assay has the potential to replace the established GenoType NTM product family in identification of culture-enriched mycobacteria. However, our research results suggest that the assay performs suboptimally and may not be feasible for use in all clinical settings.
KEYWORDS: mycobacteria, NTM, FluoroType
INTRODUCTION
Non-tuberculous mycobacteria (NTM) encompass a group of nearly 200 mycobacterial species other than Mycobacterium tuberculosis complex and Mycobacterium leprae responsible for tuberculosis (TB) and leprosy, respectively (1). Unlike M. tuberculosis, an obligate human pathogen with well-established clinical implications, NTM are opportunistic pathogens widely present in the environment (2). These characteristics inherently complicate their clinical and epidemiological assessment, as the disease spectrum can range from transient or persistent respiratory colonization to skin and soft tissue infections and chronic pulmonary diseases that closely resemble tuberculosis clinically (3). Apart from differentiating them from M. tuberculosis complex, precise identification of closely related NTM species is essential for evaluating their clinical significance and ensuring appropriate targeted treatment (4, 5).
Although M. tuberculosis complex can, in many cases, be detected directly from clinical samples using rapid molecular diagnostic assays, NTM detection and species identification predominantly relies on culture-enriched isolates. With the increasing global incidence rates of NTM over the past few decades, there is a growing interest in laboratory diagnostic workflows that enable a more accurate and optimized NTM species identification (6). In Finland, both M. tuberculosis and NTM species are notifiable to the National Infectious Disease Register and with a low incidence rate for TB, more NTM cases than M. tuberculosis cases are being reported (7).
When considering culture-based approaches, both M. tuberculosis complex and NTM follow a similar workflow that begins with the decontamination of nonsterile sample types and proceeds with culturing on mycobacteria-specific solid and liquid media (8, 9). These methods include biochemical tests, which have gradually been replaced by molecular diagnostic assays such as the GenoType product family of DNA strip hybridization assays (Hain LifeScience, Nehren, Germany), and more recently also by MALDI-TOF (matrix-assisted laser-desorption-ionization time-of-flight) (10, 11). Given the global relevance of mixed NTM infections and NTM/M. tuberculosis co-infections (12), identification methods need improvement to identify samples containing multiple species. This can be achieved using whole genome sequencing techniques, which typically serve as reference methods for NTM species identification (13).
The recently introduced FluoroType Mycobacteria assay (Hain LifeScience, Nehren, Germany) is a molecular diagnostic assay analogous to the established GenoType system. The assay allows the identification of M. tuberculosis complex and 32 NTM species/subspecies, including the differentiation of technically challenging Mycobacterium abscessus subspecies (abscessus, bolletii, and massiliense) and Mycobacterium intracellulare subspecies chimaera (referred hereafter to as M. chimaera as by the GenoType/FluoroType assays). The assay combines the coverage of the three GenoType NTM product family assays in a single assay significantly simplifying the species identification workflow (Table 1). The FluoroType Mycobacteria assay is based on LiquidArray technology combining asymmetric PCR and lights-on/lights-off probes to produce species-specific fluorescence signatures (14, 15). The assay utilizes the FluoroCycler PCR instrument for automated PCR detection and result analysis in contrast to the manual workflow and subjective result analysis of the GenoType assays.
TABLE 1.
Mycobacterial species/subspecies identified by GenoType NTM product family and FluoroType Mycobacteria assays as declared by the manufacturer
| Species/subspecies | GenoType Mycobacterium CM | GenoType Mycobacterium AS | GenoType NTM-DR | GenoType (total) | FluoroType Mycobacteria |
|---|---|---|---|---|---|
| M. abscessus complex | x | x | |||
| M. abscessus subsp. abscessus | x | x | x | ||
| M. abscessus subsp. bolletii | x | x | x | ||
| M. abscessus subsp. massiliense | x | x | x | ||
| M. asiaticum | x | x | x | ||
| M. avium | x | x | x | x | |
| M. celatum | x | x | x | ||
| M. chelonae | x | x | x | x | |
| M. chimaera | x | x | x | ||
| M. fortuitum | xa | x | xa | ||
| M. gastri | x | x | x | ||
| M. genavense | x | x | x | ||
| M. goodii | x | x | x | ||
| M. gordonae | x | x | x | ||
| M. haemophilum | x | x | x | ||
| M. heckeshornense | x | x | x | ||
| M. intermedium | x | x | x | ||
| M. interjectum | x | x | x | ||
| M. intracellulare | x | x | x | x | |
| M. kansasii | x | x | x | x | |
| M. lentiflavum | x | x | x | ||
| M. malmoense | x | x | x | ||
| M. marinum | xb | x | x | ||
| M. mucogenicum | x | x | x | ||
| M. peregrinum | x | ||||
| M. phlei | x | x | x | ||
| M. scrofulaceum | x | x | x | ||
| M. shimoidei | x | x | x | ||
| M. simiae | x | x | x | ||
| M. smegmatis | x | x | x | ||
| M. szulgai | x | x | x | x | |
| M. tuberculosis complex | x | x | x | ||
| M. ulcerans | xb | x | x | x | |
| M. xenopi | x | x | x | ||
| M. sp. (genus) | x | x | x | x | |
| Total | 14b | 16 | 7 | 32 | 33 |
Reported as M. fortuitum group by the GenoType CM assay and M. fortuitum by the FluoroType assay.
No differentiation between M. marinum and M. ulcerans (counted as one species).
The objective of this study was to assess the clinical performance of the FluoroType Mycobacteria assay in identification of mycobacterial species. Additionally, the assay was compared to the established GenoType Mycobacterium CM assay as the initial identification method, with expectations for more accurate and robust identification results, as well as an improved laboratory workflow with reduced reliance on subsequent analyses.
MATERIALS AND METHODS
Clinical mycobacterial cultures
The study material comprised 191 clinical mycobacterial cultures (53 in Löwenstein Jensen, LJ, and 138 in mycobacteria growth indicator tube, MGIT) originally cultured from 173 respiratory (sputum, bronchial washing, bronchoalveolar lavage) and 18 non-respiratory routine samples (tissue, abscess, urine, feces) in the HUS Diagnostic Center, Bacteriology laboratory, Helsinki, Finland, between March 2022 and April 2023. Of the 191 cultures, 180 were identified for single mycobacterial species, 6 for multiple mycobacterial species, and 5 for non-mycobacterial species (Table 2). All cultures were positive for acid-fast bacilli in auramine staining.
TABLE 2.
Summary of samples analyzed in the study
| Number of mycobacterial species in a sample | Clinical cultures | Other samples | Total |
|---|---|---|---|
| Single mycobacterial sp. | 180 | 180 | |
| Multiple mycobacterial sp. | 6 | 24a | 30 |
| No mycobacterial sp. | 5 | 17b | 22 |
| Total | 191 |
Nucleic acid eluates of identified clinical cultures mixed in 1:1 ratio.
Reference strains and clinical isolates identified with MALDI-TOF.
The original samples for the mycobacterial culture had been cultured at 37°C using LJ agar (Bio-Rad Laboratories, Hercules, CA, USA) or the BD BACTEC MGIT System (BD, Franklin Lakes, NJ, USA), and analyzed using the GenoType Mycobacterium CM assay (Hain LifeScience, Nehren, Germany) in the HUS Diagnostic Center. In cases where the results were equivocal, i.e., genus level identification or ambiguous banding pattern, further analysis was conducted by the national reference laboratory (Finnish Institute of Health and Welfare, Helsinki, Finland) using the GenoType Mycobacterium AS assay (Hain LifeScience, Nehren, Germany), GenoType NTM-DR assay (Hain LifeScience, Nehren, Germany), and, if necessary, 269 bp sequencing of the 16S rRNA hypervariable region. Sequencing was based on the protocol described by Kirschner et al. (16). All other laboratory works in the study were carried out at the HUS Diagnostic Center, Bacteriology Unit (Helsinki, Finland).
Clinical mycobacterial cultures were subcultured in LJ agar or MGIT medium before downstream analyses to obtain fresh cultures. Once positive, the cultures were withdrawn from incubator to room temperature and proceeded to species identification analysis within a week.
FluoroType Mycobacteria evaluation
FluoroType Mycobacteria 1.0 assay (Hain LifeScience, Nehren, Germany) was used to analyze both clinical mycobacterial cultures and non-mycobacterial control isolates. Analyses were carried out according to the manufacturer’s instructions for use (IFU) if not stated otherwise. Sample preparation was performed manually using the FluoroLyse extraction kit (Hain LifeScience, Nehren, Germany). Nucleic acid eluates were stored at −20°C for up to 2 weeks before PCR (non-IFU compliant) instead of immediate analysis. These eluates were used for all nucleic acid analyses. The FluoroType Mycobacteria assay was set up manually, and amplification and result analysis were performed with the FluoroCycler XT instrument (Hain LifeScience, Nehren, Germany). Samples with invalid results were reanalyzed.
Samples with discordant FluoroType Mycobacteria results compared to routine methods were reanalyzed using the GenoType Mycobacterium CM assay and 525 bp partial 16S rRNA sequencing (17, 18). Discordant results were resolved according to the 16S rRNA sequencing result. All samples giving M. abscessus or M. chimaera identification were analyzed using the GenoType NTM-DR assay (Hain LifeScience, Nehren, Germany). Apart from using FluoroLyse-prepared nucleic acid eluates (non-IFU compliant) instead of the analogous GenoLyse protocol, all GenoType verification analyses were conducted according to manufacturer’s instructions.
Negative control material
In addition to the original clinical mycobacterial cultures identified for non-mycobacterial species, clinical isolates and reference strains of aerobic actinomycetes and other non-mycobacterial gram-positive bacterial species (n = 17) were used to serve as additional negative control material in the study and to extend the manufacturer's evaluation of analytical specificity. Non-mycobacterial species were cultured on horse blood or chocolate agar and were identified using MALDI-TOF with Vitek MS Prime (Knowledge Base 3.0) (bioMérieux, Marcy-l’Étoile, France).
Mixed samples
In addition to the original clinical mycobacterial cultures identified for multiple species, a subset of 24 samples mimicking mixed cultures was prepared for evaluating the assay’s ability to detect multiple species in a single sample. For this, nucleic acid eluates of 48 single species cultures with valid FluoroType identification results were mixed in a 1:1 ratio. This follow-up analysis was performed using the FluoroType Mycobacteria assay only.
In silico identification analysis of register data
The mycobacterial cultures that underwent mycobacterial species identification at the HUS Diagnostic Center between 2016 and 2022 were listed using WHONET 2022 software. In accordance with the HUS Diagnostic Center mycobacterial species identification criteria, clinical cases with no species identification for at least 3 months prior to sample collection were included in the data collection.
The register data were used to estimate species identification success rates between GenoType Mycobacterium CM and FluoroType Mycobacterium in silico if used as the primary identification method.
Comparison was based on qualitative identification results stated in the assay manuals. An additional estimation was based on qualitative FluoroType Mycobacteria identification results adjusted by suboptimal (<100%) species-specific success rates observed in the study’s clinical evaluation. When using observed success rates to calculate estimations, inadequate identification results for species with only one analyzed isolate (i.e., n = 1 with 0% success rate) were excluded from in silico analysis to reduce statistical bias.
Statistical analysis
Descriptive statistics (positive and negative agreement) were calculated using Microsoft Excel and 95% confidence intervals (CI) by Wald method using GraphPad Quickcalcs.
RESULTS
FluoroType Mycobacteria analysis of clinical mycobacterial cultures and control isolates
Of the 191 clinical mycobacterial cultures analyzed in the study, 180 were identified for one mycobacterial species, 6 for multiple mycobacterial species, and 5 for non-mycobacterial species. The results of the FluoroType Mycobacteria assay evaluation are presented in Tables 3 and 4. The overall positive (PPA) and negative percent agreement (NPA) for the assay were 87.8% (95% CI: 78.4%–89.1%) and 99.6% (95% CI: 99.5%–99.8%), respectively. There were five cases of false species identification and 17 cases of insufficient, genus level identification (genus instead of species). Inadequate results (genus instead of species or false species) were observed for Mycobacterium avium (1/33), Mycobacterium branderii (1/1), Mycobacterium gordonae (9/18), Mycobacterium interjectum (3/5), Mycobacterium intracellulare (7/12), and Mycobacteriu mageritense (1/1) (Table 5). The FluoroType Mycobacteria assay further specified the GenoType CM result in 21% of cases (n = 37) (genus to species or species to subspecies). In two cases, a discordant result between GenoType CM and FluoroType Mycobacteria assays revealed an incorrect GenoType CM identification (Table 5).
TABLE 3.
Comparison of FluoroType Mycobacteria results with reference identification, organized by reference identificationd
| Reference identification | FluoroType Mycobacterium results (expected identification/total) |
Expected FluoroType Mycobacteria identificationa | |||
|---|---|---|---|---|---|
| Total | LJ | MGIT | % | ||
| Mtb complex | 29/29 | 7/7 | 22/22 | 100 | |
| M. tuberculosis | 25/25 | 4/4 | 21/21 | 100 | M. tuberculosis complex |
| M. bovis (BCG) | 4/4 | 3/3 | 1/1 | 100 | M. tuberculosis complex |
| NTM | 129/151 | 34/43 | 95/108 | 85 | |
| M. abscessus subsp. abscessus | 8/8 | 3/3 | 5/5 | 100 | M. abscessus subsp. abscessus |
| M. abscessus subsp. massiliense | 6/6 | 1/1 | 5/5 | 100 | M. abscessus subsp. massiliense |
| M. asiaticum | 1/1 | 1/1 | 100 | M. asiaticum | |
| M. avium | 32/33 | 9/9 | 23/24 | 97 | M. avium |
| M. bohemicum | 4/4 | 4/4 | 100 | M. sp. | |
| M. branderii | 0/1 | 0/1 | 0 | M. sp. | |
| M. chelonae | 3/3 | 3/3 | 100 | M. chelonae | |
| M. chimaera | 8/8 | 3/3 | 5/5 | 100 | M. chimaera |
| M. farcinogenes/senegalense/conceptionenseb | 1/1 | 1/1 | 100 | M. fortuitum | |
| M. fortuitum | 11/11 | 3/3 | 8/8 | 100 | M. fortuitum |
| M. gordonae | 9/18 | 0/3 | 9/15 | 50 | M. gordonae |
| M. immunogenumc | 1/1 | 1/1 | 100 | M. chelonae | |
| M. interjectum | 2/5 | 2/4 | 0/1 | 40 | M. interjectum |
| M. intracellulare | 5/12 | 0/2 | 5/10 | 42 | M. intracellulare |
| M. kansasii | 4/4 | 3/3 | 1/1 | 100 | M. kansasii |
| M. lentiflavum | 7/7 | 2/2 | 5/5 | 100 | M. lentiflavum |
| M. mageritense | 0/1 | 0/1 | 0 | M. sp. | |
| M. malmoense | 7/7 | 1/1 | 6/6 | 100 | M. malmoense |
| M. marinum | 1/1 | 1/1 | 100 | M. marinum | |
| M. minnosotensis | 1/1 | 1/1 | 100 | M. sp. | |
| M. mucogenicum | 2/2 | 1/1 | 1/1 | 100 | M. mucogenicum |
| M. nebraskense | 3/3 | 3/3 | 100 | M. sp. | |
| M. scrofulaceum | 1/1 | 1/1 | 100 | M. scrofulaceum | |
| M. shimoidei | 3/3 | 1/1 | 2/2 | 100 | M. shimoidei |
| M. simiae | 1/1 | 1/1 | 100 | M. simiae | |
| M. szulgai | 1/1 | 1/1 | 100 | M. szulgai | |
| M. terrae | 3/3 | 3/3 | 100 | M. sp. | |
| M. xenopi | 1/1 | 1/1 | 100 | M. xenopi | |
| M. sp (unidentified) | 3/3 | 3/3 | 100 | M. sp. | |
| Mycobacteria, total | 158/180 | 41/50 | 117/130 | 88 | |
As declared by the manufacturer.
Member of the M. fortuitum group.
Genetically related to M. abscessus and M. immunogenum.
Only clinical mycobacterial cultures identified for single mycobacterial species are listed. Discrepant results are presented in detail in Table 5.
TABLE 4.
Comparison of FluoroType Mycobacteria results with reference identification, organized by FluoroType Mycobacteria result categorya
| FluoroType Mycobacteria identification | True positive (LJ + MGIT) |
True negative (LJ + MGIT) |
False positive (LJ + MGIT) |
False negative (LJ + MGIT) |
Positive agreement (95% CI) | Negative agreement (95% CI) |
|---|---|---|---|---|---|---|
| M. abscessus subsp. abscessus | 8 (3 + 5) | 177 (48 + 129) | 100 (62.8, 100) | 100 (97.4, 100) | ||
| M. abscessus subsp. bolletii | 185 (51 + 134) | 100 (97.6, 100) | ||||
| M. abscessus subsp. massiliense | 6 (1 + 5) | 179 (50 + 129) | 100 (55.7, 100) | 100 (97.5, 100) | ||
| M. asiaticum | 1 (0 + 1) | 184 (51 + 133) | 100 (16.8, 100) | 100 (97.5, 100) | ||
| M. avium | 32 (9 + 23) | 151 (41 + 110) | 1 (1 + 0) | 1 (0 + 1) | 97.0 (83.4, 100) | 99.3 (96.0, 100) |
| M. celatum | 184 (50 + 134) | 1 (1 + 0) | 99.5 (96.7, 100) | |||
| M. chelonae | 4 (0 + 4) | 181 (51 + 130) | 100 (45.4, 100) | 100 (97.5, 100) | ||
| M. chimaera | 8 (3 + 5) | 177 (48 + 129) | 100 (62.8, 100) | 100 (97.4, 100) | ||
| M. fortuitum | 12 (3 + 9) | 173 (48 + 125) | 100 (71.8, 100) | 100 (97.4, 100) | ||
| M. gastri | 185 (51 + 134) | 100 (97.6, 100) | ||||
| M. genavense | 185 (51 + 134) | 100 (97.6, 100) | ||||
| M. goodii | 185 (51 + 134) | 100 (97.6, 100) | ||||
| M. gordonae | 9 (0 + 9) | 167 (48 + 119) | 9 (3 + 6) | 50 (29.0, 71.0) | 100 (97.3, 100) | |
| M. haemophilum | 185 (51 + 134) | 100 (97.6, 100) | ||||
| M. heckeshornense | 185 (51 + 134) | 100 (97.6, 100) | ||||
| M. intermedium | 185 (51 + 134) | 100 (97.6, 100) | ||||
| M. interjectum | 2 (2 + 0) | 180 (47 + 133) | 3 (2 + 1) | 40 (11.6, 77.1) | 100 (97.5, 100) | |
| M. intracellulare | 5 (0 + 5) | 173 (49 + 124) | 7 (2 + 5) | 41.7 (19.3, 68.1) | 100 (97.4, 100) | |
| M. kansasii | 4 (3 + 1) | 181 (48 + 133) | 100 (45.4, 100) | 100 (97.5, 100) | ||
| M. lentiflavum | 7 (2 + 5) | 178 (49 + 129) | 100 (59.6, 100) | 100 (97.5, 100) | ||
| M. malmoense | 7 (1 + 6) | 178 (50 + 128) | 100 (59.6, 100) | 100 (97.5, 100) | ||
| M. marinum | 1 (1 + 0) | 184 (50 + 134) | 100 (16.8, 100) | 100 (96.7, 100) | ||
| M. mucogenicum | 2 (1 + 1) | 183 (50 + 133) | 100 (29.0, 100) | 100 (97.5, 100) | ||
| M. peregrinum | 185 (51 + 134) | 100 (97.6, 100) | ||||
| M. phlei | 185 (51 + 134) | 100 (97.6, 100) | ||||
| M. scrofulaceum | 1 (1 + 0) | 184 (50 + 134) | 100 (16.8, 100) | 100 (96.7, 100) | ||
| M. shimoidei | 3 (1 + 2) | 181 (50 + 131) | 1 (0 + 1) | 100 (28.9, 96.6) | 99.5 (96.6, 100) | |
| M. simiae | 1 (0 + 1) | 184 (51 + 133) | 100 (16.8, 100) | 100 (96.7, 100) | ||
| M. smegmatis | 183 (50 + 133) | 2 (1 + 1) | 98.9 (95.9, 100) | |||
| M. szulgai | 1 (0 + 1) | 184 (51 + 133) | 100 (16.8, 100) | 100 (96.7, 100) | ||
| M. tuberculosis complex | 29 (7 + 22) | 156 (44 + 112) | 100 (86.1, 100) | 100 (97.1, 100) | ||
| M. ulcerans | 185 (51 + 134) | 100 (97.6, 100) | ||||
| M. xenopi | 1 (0 + 1) | 184 (51 + 133) | 100 (16.8, 100) | 100 (96.7, 100) | ||
| M. sp. (genus) | 14 (3 + 11) | 152 (40 + 122) | 17 (6 + 11) | 2 (2 + 0) | 87.5 (62.7, 97.8) | 89.9 (84.4, 93.7) |
| Total | 158 | 6,088 | 22 | 22 | 87.8 (78.4, 89.1) | 99.6 (99.5, 99.8) |
Each analyzed culture represents 34 different positive/negative FluoroType Mycobacteria identifications. All clinical mycobacterial cultures other than those identified for multiple species are included (n = 185).
TABLE 5.
Discrepant GenoType CM/FluoroType identification resultsb
| Medium | n | GenoType CM | FluoroType | 16S rRNA sequencing |
|---|---|---|---|---|
| MGIT | 1 | M. szulgai | M. sp. | M. bohemicum |
| MGIT | 1 | M. scrofulaceum | M. sp. | M. nebraskense |
| MGIT | 1 | M. avium | M. sp. | M. avium |
| LJ | 1 | M. sp. | M. celatum | M. branderii |
| LJ | 3 | M. gordonae | M. sp. | M. gordonae |
| MGIT | 6 | M. gordonae | M. sp. | M. gordonae |
| LJ | 1 | M. fortuitum group | M. smegmatis | M. mageritense a |
| MGIT | 1 | M. interjectum | M. smegmatis | M. interjectum |
| LJ | 2 | M. interjectum | M. sp. | M. interjectum |
| LJ | 1 | M. intracellulare | M. avium | M. intracellulare |
| MGIT | 1 | M. intracellulare | M. shimoidei | M. intracellulare |
| LJ | 1 | M. intracellulare | M. sp. | M. intracellulare |
| MGIT | 4 | M. intracellulare | M. sp. | M. intracellulare |
Thirteen GenoType CM identification results were complicated due to ambiguous individual band intensities. This created a risk of subjective interpretation between two different band patterns such as mycobacterial species vs non-mycobacterial species (genus control band). All 13 isolates were identified correctly with the FluoroType Mycobacteria assay.
Among the 191 clinical mycobacterial cultures, 6 were identified for multiple mycobacterial species. The presence of multiple species was initially suspected based on non-specific banding patterns or patterns matching multiple species in the GenoType CM assay. These cultures were subcultured, and single species were identified individually. The FluoroType Mycobacteria assay showed inconsistent results for these cultures with three showing single species identification and three showing identification at the genus level (Table S1). To further evaluate the assay’s ability to identify mixed cultures, 48 FluoroLyse eluates correctly identified with the FluoroType Mycobacteria assay were mixed and analyzed as 24 samples simulating cultures with multiple species. None of the samples was identified as having multiple species: 12 were identified as single species, and 12 were identified at the mycobacterial genus level (Table S1).
The analytical specificity of the FluoroType Mycobacteria assay was tested using clinical mycobacterial cultures positive for non-mycobacterial growth (n = 5; 3/5 cultures identifiable for bacterial genus/species) and various gram-positive bacterial isolates (n = 17) (Table S2). The assay showed no cross-reactivity with analyzed non-target bacteria.
In silico identification analysis of register data
A total of 2,573 cultures were analyzed for mycobacterial species at the HUS Diagnostic Center between 2016 and 2022 (Table S3). Among these, 562 (21.8%) were identified as M. tuberculosis complex and 1,970 (77.8%) as NTM. 1.6% (n = 41) of cultures involved multiple species. The most frequently identified NTM species were members of the M. avium complex (MAC) (n = 886; 45.0% of NTM) and M. gordonae (n = 282; 14.3% of NTM). 1.4% (n = 38) of NTM could not be identified to the species level. Positive cultures with non-mycobacterial species were reported similarly to culture-negative samples and could not be included in the analysis.
If the GenoType Mycobacterium CM assay was used as the primary identification method, 85.8% (n = 2,207) of analyses should have produced correct identification at the species level without the need for subsequent identification by another method. If instead the FluoroType Mycobacteria assay was used as the primary method, this rate would have been expected to increase to 93.5% (n = 2,407). Additionally, the FluoroType Mycobacteria assay offered the added value of simultaneous identification of M. abscessus subspecies and M. chimaera. These estimations were based on the expected identification results declared in the assay manuals. However, when the expected FluoroType Mycobacterium results were adjusted according to the observed evaluation data including the suboptimal (<100%) identification success rates for M. avium, M. gordonae, M. interjectum, and M. intracellulare, the assay was estimated to produce 81.7% (n = 2,102) of identifications correctly at the species level if used as the primary identification method. This corresponded to an estimated 29% (n = 105) increase in inadequate identification results (genus or false species) compared to the use of the GenoType CM assay as the primary identification method.
DISCUSSION
The FluoroType Mycobacteria assay is a molecular diagnostic assay designed for the identification of M. tuberculosis complex and 32 different NTM species/subspecies, similar to the GenoType NTM product family. Compared to the manual GenoType assays, the FluoroType Mycobacteria assay offers technical ease and significantly reduces labor and time requirements. This efficiency is primarily attributed to the FluoroCycler PCR instrument, which automates the stages of target DNA detection and result analysis. Additionally, the assay eliminates the need for handling amplified PCR products, reducing the risk of amplicon cross-contamination. On the contrary, the GenoType workflow does not rely on an expensive or specialized instrument and can be carried out using a generic thermal cycler.
In our laboratory setting, where mycobacterial species identification represents only a fraction of the overall workload, it is important for a hardware-dependent identification method to synergize with other clinical laboratory processes. This is where the FluoroType Mycobacteria assay’s closed system design, dependent on the FluoroCycler PCR instrument, presents a clear disadvantage, although there are some other assays able to be used in the same platform. Currently, the assay has only been validated for manual sample preparation. During our evaluation, we analyzed a small subset of samples (n = 9) prepared using the automatic MagNA Pure 24 System (Roche Molecular Systems, Pleasanton, CA, USA) along with the manual protocol. The results showed no differences between the two sample preparation methods (results not shown), indicating potential for incorporating the assay into an existing automated workflow. This would significantly enhance the assay’s flexibility and move it closer to full automation.
One of the most evident advantages of the FluoroType Mycobacteria assay, compared to the GenoType product family, is its ability to cover a wider repertoire of species with a single assay, thereby theoretically reducing the need for subsequent analyses by other identification methods. Ideally, this feature would partially compensate for the initial investment costs associated with the FluoroCycler PCR instrument. However, based on the clinical performance data observed in this study, this advantage would be likely to be diminished due to the significant proportion of common species, particularly M. gordonae, being identified as Mycobacterium sp. Globally, M. gordonae is the second or third most frequently isolated NTM species, with reported distribution rates of up to 29% among locally isolated NTM species (19). Overall, in this study’s clinical setting, the use of the FluoroType Mycobacteria assay, in fact, would have increased the number of subsequent analyses by secondary identification methods. However, although this is a technical weakness, it mainly affects the identification process expenses while having limited significance in terms of patient care. This effect may be lesser in clinical settings with lower incidence rates for M. gordonae and M. intracellulare.
As an assay offering automatic result analysis and a broader range of identified species, the FluoroType Mycobacteria assay was expected to address the GenoType assay’s challenges with subjective result interpretation. Specifically, the assay would be expected to overcome issues related to faint background banding patterns and patterns representing multiple species (20). The FluoroType Mycobacteria assay successfully resolved all cases where GenoType result interpretation was complicated by ambiguous band intensity. This highlights one of the primary advantages of automatic and objective result interpretation. However, with samples containing multiple species, the FluoroType Mycobacteria assay did not provide any improvement. The assay would either provide a genus identification or identify a single species without indicating the presence of multiple species. Since the assay technology is based on detecting species-specific melting temperature curves, a mixed identification would simply be interpreted as the genus Mycobacterium without a species-specific curve in the database. Therefore, the assay is practically unable to suggest the presence of multiple species in a single sample, although this is not clearly stated by the manufacturer. Importantly, however, even though this feature poses an evident risk of diagnostic bias, the clinically significant species, such as M. tuberculosis, M. abscessus, and Mycobacterium kansasii, were generally identified in mixed samples and were not obscured by identification of less relevant mycobacterial species. This phenomenon may be partly attributed to competition during amplification, although no consistent pattern was observed with regard to the species-specific limit of detection (LOD) levels stated by the manufacturer. These vary substantially between species: from 47 CFU/mL (for M. tuberculosis) to 257,040 CFU/mL (for M. abscessus subsp. abscessus) (median = 5,260.5 CFU/mL, n = 30).
There were a few limitations in this study. At the study site, the GenoType Mycobacterium CM assay is the primary, and in most cases, the sole method used to achieve sufficient level of NTM species identification. Consequently, the register data (mycobacterial species reported between 2016 and 2022) used to compare the coverage of the GenoType and FluoroType Mycobacteria assays may be biased. For example, in the workflow, M. chimaera was only suspected based on atypical M. intracellulare banding patterns in the GenoType Mycobacterium CM assay (21). This method may miss some M. chimaera isolates and underestimate the need for M. chimaera identification by the primary identification test (22). The same applies to cultures identified for Mycobacterium malmoense or Mycobacterium scrofulaceum. Similarly, it should be noted that in some cases, accurate identification of mycobacterial species might require nucleic acid sequence analysis not restricted to partial 16S rRNA sequence. In case of M. gordonae, for example, identification between M. gordonae and the more recently described M. gordonae complex species Mycobacterium paragordonae or Mycobacterium vicinigordonae should be based on additional sequencing targets such as hsp65 and rpoB for reliable resolution (23, 24). The same applies to distinguishing M. intracellulare from some other MAC species such as Mycobacterium marseillense and Mycobacterium timonense (25). It is thus possible that some of the inadequately identified cultures (genus instead of species), in fact, represent a closely related species or a distinct subspecies. Interestingly, however, the manufacturer declares that the FluoroType assay reports most of the Mycobacterium fortuitum group species as M. fortuitum or Mycobacterium peregrinum instead of Mycobacterium sp. Finally, the 29 mycobacterial species analyzed in this study represented only approximately 15% of the currently identified approximately 200 mycobacterial species. Although these 29 species covered over 97% of the isolates identified at the study site between 2016 and 2022, further evaluation showing identification results for other clinically rare species would be important. It is known that successful treatment of NTM infections can be species-specific, and thus, false species identification for rare species (any result other than Mycobacterium sp.) could potentially lead to incorrect treatment (3–5).
In conclusion, the FluoroType Mycobacteria assay exhibited some advantageous features in comparison to the GenoType Mycobacterium CM assay including wider range of identified species, discrimination of the M. abscessus subspecies and M. intracellulare subspecies chimaera, and improvements in process workflow and objective result interpretation. However, it also had evident weaknesses in identifying mixed cultures and some commonly encountered species such as M. intracellulare and M. gordonae, which may, in fact, increase the need for subsequent analyses by other identification methods. Therefore, the cost-efficiency of the FluoroType Mycobacteria assay should be assessed with respect to the local NTM species distribution to estimate the feasibility of the assay in a clinical setting.
ACKNOWLEDGMENTS
We thank Anne-Maj Hurnanen, Mira Id, Fanni Jarske, and other Mycobacteriology and Nucleic Acid laboratory personnel of HUS Diagnostic Center for essential contribution. We also thank Immuno Diagnostic Ltd. and Hain LifeScience GmbH. Immuno Diagnostic or Hain LifeScience did not have influence in the study content.
The study was funded by the HUS Diagnostic Center.
B.L. contributed to the study design, result analysis, and drafting of the manuscript. J.An., J.Ai., and A.P-S. contributed to the supervision of the study and revision of the manuscript. T.M-L., A.P-S., and H-L.H. contributed to the original result analysis. All authors have reviewed the manuscript and provided significant input.
Contributor Information
Bruno Luukinen, Email: bruno.luukinen@fimlab.fi.
Christine Y. Turenne, University of Manitoba, Winnipeg, Manitoba, Canada
SUPPLEMENTAL MATERIAL
The following material is available online at https://doi.org/10.1128/jcm.01054-23.
Data for mixed cultures, non-mycobacterial control species, and in register data analysis.
ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.
REFERENCES
- 1. Parte AC, Sardà Carbasse J, Meier-Kolthoff JP, Reimer LC, Göker M. 2020. List of prokaryotic names with standing in nomenclature (LPSN) moves to the DSMZ. Int J Syst Evol Microbiol 70:5607–5612. doi: 10.1099/ijsem.0.004332 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Falkinham JO. 2015. Environmental sources of nontuberculous mycobacteria. Clin Chest Med 36:35–41. doi: 10.1016/j.ccm.2014.10.003 [DOI] [PubMed] [Google Scholar]
- 3. Griffith DE, Aksamit T, Brown-Elliott BA, Catanzaro A, Daley C, Gordin F, Holland SM, Horsburgh R, Huitt G, Iademarco MF, Iseman M, Olivier K, Ruoss S, von Reyn CF, Wallace RJ Jr, Winthrop K, ATS Mycobacterial Diseases Subcommittee, American Thoracic Society, Infectious Disease Society of America . 2007. An official ATS/IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases. Am J Respir Crit Care Med 175:367–416. doi: 10.1164/rccm.200604-571ST [DOI] [PubMed] [Google Scholar]
- 4. Daley CL, Iaccarino JM, Lange C, Cambau E, Wallace RJ Jr, Andrejak C, Böttger EC, Brozek J, Griffith DE, Guglielmetti L, Huitt GA, Knight SL, Leitman P, Marras TK, Olivier KN, Santin M, Stout JE, Tortoli E, van Ingen J, Wagner D, Winthrop KL. 2020. Treatment of nontuberculous mycobacterial pulmonary disease: an official ats/ers/escmid/idsa clinical practice guideline. Clin Infect Dis 71:e1–e36. doi: 10.1093/cid/ciaa241 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Lange C, Böttger EC, Cambau E, Griffith DE, Guglielmetti L, van Ingen J, Knight SL, Marras TK, Olivier KN, Santin M, Stout JE, Tortoli E, Wagner D, Winthrop K, Daley CL. 2022. Consensus management recommendations for less common non-tuberculous mycobacterial pulmonary diseases. Lancet Infect Dis 22:e178–e190. doi: 10.1016/S1473-3099(21)00586-7 [DOI] [PubMed] [Google Scholar]
- 6. Ratnatunga CN, Lutzky VP, Kupz A, Doolan DL, Reid DW, Field M, Bell SC, Thomson RM, Miles JJ. 2020. The rise of non-tuberculosis mycobacterial lung disease. Front Immunol 11:303. doi: 10.3389/fimmu.2020.00303 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Finnish Institute for Health and Welfare . 2023. {Data set.}. Statistical database of the national infectious diseases register (in Finnish and Swedish). Available from: https://www.thl.fi/ttr/gen/rpt/tilastot.html
- 8. Dorman SE, Schumacher SG, Alland D, Nabeta P, Armstrong DT, King B, Hall SL, Chakravorty S, Cirillo DM, Tukvadze N, et al. 2018. Xpert MTB/RIF ultra for detection of Mycobacterium tuberculosis and rifampicin resistance: a prospective multicentre diagnostic accuracy study. Lancet Infect Dis 18:76–84. doi: 10.1016/S1473-3099(17)30691-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Hofmann-Thiel S, Molodtsov N, Antonenka U, Hoffmann H. 2016. Evaluation of the Abbott RealTime MTB and RealTime MTB INH/RIF assays for direct detection of Mycobacterium tuberculosis complex and resistance markers in respiratory and extrapulmonary specimens. J Clin Microbiol 54:3022–3027. doi: 10.1128/JCM.01144-16 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Mäkinen J, Marjamäki M, Marttila H, Soini H. 2006. Evaluation of a novel strip test, GenoType Mycobacterium CM/AS, for species identification of mycobacterial cultures. Clin Microbiol Infect 12:481–483. doi: 10.1111/j.1469-0691.2006.01380.x [DOI] [PubMed] [Google Scholar]
- 11. Rodríguez-Sánchez B, Ruiz-Serrano MJ, Ruiz A, Timke M, Kostrzewa M, Bouza E. 2016. Evaluation of MALDI biotyper mycobacteria library v3.0 for identification of nontuberculous mycobacteria. J Clin Microbiol 54:1144–1147. doi: 10.1128/JCM.02760-15 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Byrne AS, Goudreau A, Bissonnette N, Shamputa IC, Tahlan K. 2020. Methods for detecting mycobacterial mixed strain infections–A systematic review. Front Genet 11:600692. doi: 10.3389/fgene.2020.600692 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Dohál M, Porvazník I, Solovič I, Mokrý J. 2021. Whole genome sequencing in the management of non-tuberculous mycobacterial infections. Microorganisms 9:11. doi: 10.3390/microorganisms9112237 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Sanchez JA, Pierce KE, Rice JE, Wangh LJ. 2004. Linear-after-the-exponential (LATE)-PCR: an advanced method of asymmetric PCR and its uses in quantitative real-time analysis. Proc Natl Acad Sci U S A 101:1933–1938. doi: 10.1073/pnas.0305476101 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Rice JE, Reis AH, Rice LM, Carver-Brown RK, Wangh LJ. 2012. Fluorescent signatures for variable DNA sequences. Nucleic Acids Res 40:21. doi: 10.1093/nar/gks731 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Kirschner P, Springer B, Vogel U, Meier A, Wrede A, Kiekenbeck M, Bange FC, Böttger EC. 1993. Genotypic identification of mycobacteria by nucleic acid sequence determination: report of a 2-year experience in a clinical laboratory. J Clin Microbiol 31:2882–2889. doi: 10.1128/jcm.31.11.2882-2889.1993 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Edwards KJ, Logan JMJ, Langham S, Swift C, Gharbia SE. 2012. Utility of real-time amplification of selected 16S rRNA gene sequences as a tool for detection and identification of microbial signatures directly from clinical samples. J Med Microbiol 61:645–652. doi: 10.1099/jmm.0.041764-0 [DOI] [PubMed] [Google Scholar]
- 18. Halavaara M, Martelius T, Järvinen A, Antikainen J, Kuusela P, Salminen US, Anttila VJ. 2019. Impact of pre-operative antimicrobial treatment on microbiological findings from endocardial specimens in infective endocarditis. Eur J Clin Microbiol Infect Dis 38:497–503. doi: 10.1007/s10096-018-03451-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Hoefsloot W, Van Ingen J, Andrejak C, Ngeby K, Bauriaud R, Bemer P, Beylis N, Boeree MJ, Cacho J, Chihota V. 2013. The geographic diversity of Nontuberculous mycobacteria isolated from pulmonary samples: An NTM-NET collaborative study for the Nontuberculous mycobacteria network European trials group (NTM-NET). Eur Respir J 42:1604–1613. doi: 10.1183/09031936.00149212 [DOI] [PubMed] [Google Scholar]
- 20. Richter E, Rüsch-Gerdes S, Hillemann D. 2006. Evaluation of the GenoType mycobacterium assay for identification of mycobacterial species from cultures. J Clin Microbiol 44:1769–1775. doi: 10.1128/JCM.44.5.1769-1775.2006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Lecorche E, Haenn S, Mougari F, Kumanski S, Veziris N, Benmansour H, Raskine L, Moulin L, Cambau E, Aubry A, Brossier F, Chauffour A, Jaffre J, Jarlier V, Robert J, Sougakoff W. 2018. Comparison of methods available for identification of Mycobacterium chimaera. Clin Microbiol Infect 24:409–413. doi: 10.1016/j.cmi.2017.07.031 [DOI] [PubMed] [Google Scholar]
- 22. Zabost AT, Szturmowicz M, Brzezińska SA, Klatt MD, Augustynowicz-Kopeć EM. 2021. Mycobacterium chimaera as an underestimated cause of NTM lung diseases in patients hospitalized in pulmonary wards. Pol J Microbiol 70:315–320. doi: 10.33073/pjm-2021-028 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Kim BJ, Hong SH, Kook YH, Kim BJ. 2014. Mycobacterium paragordonae sp. nov., a slowly growing, scotochromogenic species closely related to Mycobacterium gordonae. Int J Syst Evol Microbiol 64:39–45. doi: 10.1099/ijs.0.051540-0 [DOI] [PubMed] [Google Scholar]
- 24. Liu G, Yu X, Luo J, Hu Y, Dong L, Jiang G, Huo F, Wang F, Liang Q, Huang H. 2021. Mycobacterium vicinigordonae sp. nov., a slow-growing scotochromogenic species isolated from sputum. Int J Syst Evol Microbiol 71. doi: 10.1099/ijsem.0.004796 [DOI] [PubMed] [Google Scholar]
- 25. Ben Salah I, Cayrou C, Raoult D, Drancourt M. 2009. Mycobacterium marseillense sp. nov., Mycobacterium timonense sp. nov. and Mycobacterium bouchedurhonense sp. nov., members of the Mycobacterium avium complex. Int J Syst Evol Microbiol 59:2803–2808. doi: 10.1099/ijs.0.010637-0 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data for mixed cultures, non-mycobacterial control species, and in register data analysis.