Combining COLD-PCR and high-resolution melt analysis for rapid detection of low-level, rifampin-resistant mutations in Mycobacterium tuberculosis

Dependence of Xpert MTB/RIF Accuracy for Detecting Rifampin Resistance in Bronchoalveolar Lavage Fluid on Bacterial Load: A Retrospective Study in Beijing, China

INTRODUCTION

We assessed the effect of Mycobacterium tuberculosis (MTB) bacterial load on Xpert MTB/RIF accuracy for detection of rifampicin (RIF)-resistant MTB in bronchoalveolar lavage fluid (BALF) specimens obtained at a national tuberculosis (TB) specialized hospital in Beijing, China.

METHODS

A retrospective study was conducted at Beijing Chest Hospital. Patients with symptoms suggestive of pulmonary TB who provided BALF specimens for routine MTB detection between June 2019 and July 2020 were enrolled in the study. Chi-square test and Student's t-test were used to compare results across groups stratified according to BALF bacterial load.

RESULTS

In total, 1125 patients with positive Xpert results who were enrolled in final analysis, 263 provided BALF specimens that tested positive for RIF-resistant MTB via Xpert MTB/RIF. The RIF-resistance rate of specimens with very low MTB bacterial load was 30.9%, a resistance rate significantly greater than rates obtained for groups with high (25.0%), medium (17.3%) and low (19.2%) MTB loads (P<0.01). Notably, false-positive results obtained for the very low bacterial load group led to markedly reduced positive predictive value of Xpert MTB/RIF to provide correct RIF-resistance predictions for that group (67.1%, 95% CI: 56.1%-78.1%5) relative to the predictive value obtained for all other groups combined (about 90%, P<0.05). Sanger sequencing data obtained for 20 (32.8%) MTB isolates deemed RIF-resistant via Xpert (Probe E) lacked rpoB RRDR mutations. Meanwhile, of another group of 23 isolates deemed RIF-susceptible via DST but RIF-resistant via Xpert MTB/RIF, 20 isolate sequences (87.0%) lacked rpoB RRDR mutations, while sequences of the remaining 3 isolates harbored single rpoB RRDR mutations predicted to cause amino acid substitutions.

CONCLUSION

Xpert MTB/RIF assay performed alarmingly poorly when used to detect RIF-resistant MTB in BALF specimens with very low bacterial loads. A high rate of Xpert probe E hybridization failure was the main driver of false-positive RIF-resistant results.

Assessment of current diagnostic algorithm for detection of mixed infection with Mycobacterium tuberculosis and nontuberculous mycobacteria

BACKGROUND

The increasing pulmonary diseases are reported to be affected by mixed infection of Mycobacterium tuberculosis (MTB) and nontuberculous mycobacteria (NTM). In this study, our objective was to assess the efficiency of mycobacterial culture plus DNA sequencing to detect the mixed infections with MTB and various NTM organisms. We also aimed to investigate how efficiently GeneXpert detected MTB in mixed infections with NTM in in vitro models.

METHODS

A serial of mixed infection samples was generated by combining suspensions of MTB and five NTM bacteria, respectively. The mixed suspensions were further detected with GeneXpert and liquid culture plus DNA sequencing.

RESULTS

Overall, the GeneXpert assay exhibited promising capability to identify the presence of MTB at different proportions ranging from 1% to 99%. For the liquid culture, the subsequent DNA sequencing only detected the presence of NTM bacteria in the mixed samples, which the proportion of NTM ranged from 1% to 99%, including M. intracellulare, M. kansasii, M. abscessus, and M. fortuitum. For M. avium, DNA sequencing was able to identify the mixtures as M. avium infection in suspensions with no less than 10% M. avium bacteria, whereas only MTB was found in the other suspensions with less M. avium bacteria.

CONCLUSIONS

Our data demonstrate that the current diagnostic algorithm cannot yield a precise detection of mixed infections with MTB and NTM bacteria. The GeneXpert assay only identify MTB in the mixed samples, while the subculture plus DNA sequencing prefers to identify the NTM species with the higher growth rate. Further targeted molecular analysis by specific capture of multiple loci of mycobacterial species from specimens is urgently required to solve this diagnostic dilemma.

Genomic Analysis of the Evolution of Fluoroquinolone Resistance in Mycobacterium tuberculosis Prior to Tuberculosis Diagnosis

Fluoroquinolones (FQs) are effective second-line drugs for treating antibiotic-resistant tuberculosis (TB) and are being considered for use as first-line agents. Because FQs are used to treat a range of infections, in a setting of undiagnosed TB, there is potential to select for drug-resistant Mycobacterium tuberculosis mutants during FQ-based treatment of other infections, including pneumonia. Here we present a detailed characterization of ofloxacin-resistant M. tuberculosis samples isolated directly from patients in Taiwan, which demonstrates that selection for FQ resistance can occur within patients who have not received FQs for the treatment of TB. Several of these samples showed no mutations in gyrA or gyrB based on PCR-based molecular assays, but genome-wide next-generation sequencing (NGS) revealed minority populations of gyrA and/or gyrB mutants. In other samples with PCR-detectable gyrA mutations, NGS revealed subpopulations containing alternative resistance-associated genotypes. Isolation of individual clones from these apparently heterogeneous samples confirmed the presence of the minority drug-resistant variants suggested by the NGS data. Further NGS of these purified clones established evolutionary links between FQ-sensitive and -resistant clones derived from the same patient, suggesting de novo emergence of FQ-resistant TB. Importantly, most of these samples were isolated from patients without a history of FQ treatment for TB. Thus, selective pressure applied by FQ monotherapy in the setting of undiagnosed TB infection appears to be able to drive the full or partial emergence of FQ-resistant M. tuberculosis, which has the potential to confound diagnostic tests for antibiotic susceptibility and limit the effectiveness of FQs in TB treatment.

Prevalence and risk factors of mixed Mycobacterium tuberculosis complex infections in China

OBJECTIVES

Mixed infections have been considered as a potential obstacle for tuberculosis treatment and control. To date, few studies have been done to determine the rate of mixed infections of Mycobacterium tuberculosis in China.

METHODS

In this study, we used the standard 24-loci MIRU-VNTR method to genotype the representative M. tuberculosis isolates from the national drug-resistant survey conducted in China. A total of 3248 M. tuberculosis complex (MTBC) strains had complete 24-loci MIRU-VNTR results and available for the analyses.

RESULTS

Overall, MIRU-VNTR typing identified 115 (3.5%) isolates as being mixed MTBC infections in China. Statistical analysis revealed that mixed infections were significantly more likely to occur in men than women. Compared with the percentage of mixed infection from patients aged 45-56 years, the percentages of mixed infections were higher among patients aged 25-44 years [OR (95% CI): 1.844(1.129-3.014)] and old patients [older than 65 years OR (95% CI): 1.908(1.097-3.319)]. In addition, significantly higher frequencies of hemoptysis (P = 0.022) and chest pain (P = 0.012) were observed among mixed infections, using patients infected with a single strain as a reference.

CONCLUSIONS

In conclusion, this study has provided the first comprehensive understanding of mixed MTBC infections in China, which will be essential to generate the effective TB control strategies.

Trainable high resolution melt curve machine learning classifier for large-scale reliable genotyping of sequence variants

High resolution melt (HRM) is gaining considerable popularity as a simple and robust method for genotyping sequence variants. However, accurate genotyping of an unknown sample for which a large number of possible variants may exist will require an automated HRM curve identification method capable of comparing unknowns against a large cohort of known sequence variants. Herein, we describe a new method for automated HRM curve classification based on machine learning methods and learned tolerance for reaction condition deviations. We tested this method in silico through multiple cross-validations using curves generated from 9 different simulated experimental conditions to classify 92 known serotypes of Streptococcus pneumoniae and demonstrated over 99% accuracy with 8 training curves per serotype. In vitro verification of the algorithm was tested using sequence variants of a cancer-related gene and demonstrated 100% accuracy with 3 training curves per sequence variant. The machine learning algorithm enabled reliable, scalable, and automated HRM genotyping analysis with broad potential clinical and epidemiological applications.

Comparison of different drug susceptibility test methods to detect rifampin heteroresistance in Mycobacterium tuberculosis

We compared the efficiencies of different drug susceptibility testing methods in detecting rifampin (RIF) heteroresistance in Mycobacterium tuberculosis. Our data revealed that the broth dilution method found more resistance than MGIT did (P=0.046) for the low-resistance group. Similarly, the broth dilution method was more sensitive in detecting RIF heteroresistance in subpopulations with low growth rates than was MGIT (P=0.033). In conclusion, our data demonstrated that the broth dilution method was more sensitive than MGIT in detecting RIF heteroresistance.

Molecular characterization of multidrug-resistant Mycobacterium tuberculosis isolates from China

To investigate the molecular characterization of multidrug-resistant tuberculosis (MDR-TB) isolates from China and the association of specific mutations conferring drug resistance with strains of different genotypes, we performed spoligotyping and sequenced nine loci (katG, inhA, the oxyR-ahpC intergenic region, rpoB, tlyA, eis, rrs, gyrA, and gyrB) for 128 MDR-TB isolates. Our results showed that 108 isolates (84.4%) were Beijing family strains, 64 (59.3%) of which were identified as modern Beijing strains. Compared with the phenotypic data, the sensitivity and specificity of DNA sequencing were 89.1% and 100.0%, respectively, for isoniazid (INH) resistance, 93.8% and 100.0% for rifampin (RIF) resistance, 60.0% and 99.4% for capreomycin (CAP) resistance, 84.6% and 99.4% for kanamycin (KAN) resistance, and 90.0% and 100.0% for ofloxacin (OFX) resistance. The most prevalent mutations among the MDR-TB isolates were katG315, inhA15, rpoB531, -526, and -516, rrs1401, eis-10, and gyrA94, -90, and -91. Furthermore, there was no association between specific resistance-conferring mutations and the strain genotype. These findings will be helpful for the establishment of rapid molecular diagnostic methods to be implemented in China.

High resolution melting curve assay for rapid detection of drug-resistant Mycobacterium tuberculosis

We developed and evaluated a high resolution melting (HRM) curve assay by using real-time PCR for the detection of the most frequent mutations of Mycobacterium tuberculosis, which are responsible for the resistance of four anti-TB drugs: rifampicin, isoniazid, ethambutol, and streptomycin. The HRM assay was successfully used for the detection of dominant mutations: A516V, H526A, H526T, S531L, L533P, and A516G/S531L in rpoB; S315T, and S315A in katG; -15C/T, and -8T/C in mab-inhA; M306I in embB; K88Q and K43R in rpsL; and 513A/C in rrs. We were able to discriminate the mutant from the wild type by analyzing the melting-curve shape in 40 clinical M. tuberculosis isolates, and the results of the HRM assay were completely consistent with those of DNA sequencing. This HRM assay is a simple, rapid, and cost-effective method that can be performed in a closed tube. Therefore, our assay is a potentially useful tool for the rapid detection of drug-resistant M. tuberculosis.