Linezolid phenotypic resistance in Mycobacterium tuberculosis due to presence of non-tuberculous mycobacteria:a case report

Technical Summary: Linezolid Phenotypic Resistance in Mycobacterium tuberculosis Due to Non-Tuberculous Mycobacteria

Problem Statement
Discordance between phenotypic drug-susceptibility testing (pDST) and genotypic drug-susceptibility testing (gDST) complicates clinical decision-making for rifampicin-resistant tuberculosis (RR-TB). While gDST (via whole-genome sequencing, WGS) is increasingly used, it can yield false-negative results if resistance mechanisms are unknown or if resistant subpopulations are below detection thresholds. Conversely, pDST can yield false-positive results due to technical challenges or the presence of mixed infections. This case report investigates a specific instance of linezolid discordance where pDST indicated resistance, but WGS detected no resistance-conferring mutations in the rrl or rplC genes of Mycobacterium tuberculosis (Mtb). The authors hypothesized that this discordance was caused by contamination or mixed infection with a non-tuberculous mycobacterium (NTM), given that linezolid targets ribosomal DNA, a highly conserved mechanism across bacterial species.

Methodology
The study centers on a 48-year-old male patient in South Africa with a history of multiple RR-TB episodes, diagnosed with a new episode in 2025. The investigation employed a multi-modal approach to resolve the initial pDST/gDST discordance:

• Initial Testing: Baseline isolates underwent pDST (BACTEC MGIT960) and Illumina-based WGS. A line probe assay was also performed.
• Deep Sequencing and Profiling: To investigate the discordance, the authors utilized:
  • NTM-Profiler (v0.8.2): Applied to baseline WGS data to identify species and relative abundances.
  • Genomic Alignment: Alignment of rrl and rplC genes between Mtb (H37Rv reference) and identified NTM strains (e.g., M. sinense JDM601) to assess conservation of resistance-associated positions.
  • Minimum Inhibitory Concentration (MIC) Assays: Performed on the baseline isolate and specific drug-containing MGIT tubes.
  • Targeted NGS (tNGS): The Deeplex MycMTB assay was applied to the linezolid-containing pDST tube.
  • Subculture WGS: WGS was performed on subcultures of the baseline isolate and specific growth control/drug-containing tubes to determine species composition in the presence of the drug.
• Follow-up: The patient's month-two culture isolate was subjected to Xpert MTB/XDR, WGS, and extended pDST to monitor treatment response and emerging resistance.

Key Results

• Initial Discordance: The baseline isolate showed phenotypic resistance to linezolid (MIC ≥ 1 mg/L) but genotypic susceptibility (no variants in rrl or rplC).
• Identification of Mixed Infection:
  • NTM-Profiler analysis of the baseline WGS data revealed a mixed infection: 34.6% M. tuberculosis and 65.4% Mycobacterium sinense.
  • Genomic alignment confirmed that M. sinense possesses high identity (84% for rplC, 91% for rrl) with Mtb at the seven positions associated with linezolid resistance, explaining why the NTM would appear resistant in a phenotypic assay.
• Mechanism of False Resistance:
  • WGS of the growth control tube confirmed the presence of both species.
  • Crucially, WGS of the linezolid-containing MGIT tubes (0.5 and 2.0 mg/L) detected only M. sinense, with no M. tuberculosis detected.
  • This confirmed that the M. tuberculosis strain was susceptible (it did not grow in the presence of linezolid), while the M. sinense strain was resistant and grew, leading the pDST to erroneously report the Mtb isolate as resistant.
  • The tNGS (Deeplex MycMTB) on the drug tube detected M. kumamotonense, but genomic alignment suggested the actual contaminant was M. sinense, highlighting limitations in single-marker identification methods.
• Clinical Course: The patient initially received a BPaL-L regimen. Due to the uncertainty of linezolid susceptibility and month-two culture positivity, a complex 9-drug regimen was recommended. The patient eventually achieved culture conversion. Month-two isolates confirmed Mtb susceptibility to linezolid but revealed emerging resistance to bedaquiline and clofazimine.

Significance and Claims
The paper claims to report the first case of linezolid pDST-gDST discordance where the phenotypic resistance was caused by the presence of a slow-growing NTM (M. sinense) rather than a false-negative genotypic result or a technical failure of WGS.

The authors argue that this finding challenges the prevailing assumption that linezolid discordance is primarily due to the inability of NGS to detect resistance mutations. Instead, they posit that the presence of NTM in culture isolates can lead to false-resistant pDST results. The study concludes that when sequencing results do not detect a resistance-conferring mutation, the presence of an NTM should be investigated via macroscopic examination and Ziehl-Neelsen staining before reporting a phenotypic result as Mtb linezolid-resistant. The authors suggest that understanding the frequency of this phenomenon is essential for accurately assessing the true sensitivity of NGS for linezolid resistance detection.