To respond to the urgent need for rapid and accurate detection of M. tuberculosis and drug resistance, a DNA microarray system was developed by the CapitalBio Corporation. This microarray can simultaneously identify 17 mycobacterial species by targeting the species-specific sequences in the 16S rRNA. Moreover, the specialized microarray consists of 16 oligonucleotide probes, which allowed the detection of the 13 most common mutations in the rpoB gene, 2 mutations in the katG gene, and one mutation in the promoter region of the inhA gene. In our study, the identification of M. tuberculosis strains and detection of their resistant form in spinal tuberculosis specimens were evaluated by the CapitalBio™ DNA microarray assay.
We evaluated the utility of the CapitalBio™ DNA microarray in 153 consecutive spinal tuberculosis samples. The microarray system identified more than 93% of all cases with culture-confirmed spinal tuberculosis, including more than 73% of patients with smear-negative cultures. Moreover, this assay was positive in 45% (27/60) of the patients who were not confirmed by M. tuberculosis culture but were demonstrated both clinically and pathologically to have spinal tuberculosis. Results of the study revealed that the microarray assay has a higher susceptibility compared with the conventional BACT/MGIT 960 system for the detection of M. tuberculosis strains. Previous studies conducted by Guo and Zhu et al. also reported that for the direct detection of M. tuberculosis complex bacteria the overall sensitivity obtained with this commercial DNA microarray were 100% [11, 12]. The negative control of the species identification has not been setup considering the following reasons: First, Guo and Zhu respectively found that the specificity of the CapitalBio™ DNA microarray was 100% for the species identification of M. tuberculosis[11, 12]. Second, recruitment and enrollment of pyogenic spondylitis patients, with similar clinical characteristics and clinical specimens to spinal tuberculosis, are relatively difficult in our department. However, absence of the negative control might be a potential limitation of this study.
For the detection of anti-tuberculosis drug resistance, the DNA microarray had a sensitivity of 73.3% and specificity of 91.0% for INH resistance, and a sensitivity of 88.9% and specificity of 90.7% for RMP resistance. Interestingly, 5.6% of phenotypically defined RMP-resistant specimens and 13.3% of INH- resistant specimens had no mutations in related genes, suggesting that other mechanisms or mutations in other codons of the related genes may be responsible for the emergence of RMP and INH resistance. Moreover, a small proportion of specimens failed to produce results; one possible reason for the lack of results may be because of the exiguous M. tuberculosis bacilli contained in some specimens, or the DNA amplification inhibitors in the clinical specimens, resulting in a lack of DNA being extracted from these samples. Results of the DNA microarray assay could be obtained within a mean turnaround time of 5.8 hours from the start of the analysis.
Recently, some novel molecular DST methods based on nucleic acid amplification have been available. Among them, the Xpert MTB/RIF assay was endorsed by the WHO as a replacement for sputum smear microscopy [7, 8, 14–16]. Compared with the Xpert MTB/RIF assay, the CapitalBio™ microarray system has some limitations, e.g., the assay is semi-automatic, requires more experienced technical skills for manual manipulation and has more opportunities to create biohazard intermediates. However, the microarray system has two advantages compared to the Xpert MTB/RIF assays. First, the microarray assay is less costly because cartridges of the Xpert MTB/RIF are disposable. Second, the microarray assay can detect the gene mutations associated to both RMP and INH.
In addition, compared with isolates and sputum samples, the application of the DNA microarray in spinal tuberculosis samples is far more complex because of the diversity of clinical sample types, difficulties in obtaining adequate tissue for analyses and the methods of processing samples prior to analysis .