Abstract
Tuberculosis (TB) and multidrug-resistant tuberculosis (MDR-TB) are the major public health issue in Thailand. Previously, it was found that treatment coverage rates of TB and MDR-TB in Thailand were low as a high number of TB-infected persons (>40%) were either undiagnosed or unregistered treatment. The national strategic plan (NSP) for TB (2017-2021) has been launched to eliminate TB in Thailand by 2035. To respond to the NSP, in particular, strategy 5: research and development of novel diagnostic innovation, our research aimed to evaluate the effectiveness of the molecular methods formerly developed to diagnose TB and MDR-TB in clinical samples. This study will lead to a decrease in TB diagnosis turnaround time and an increase in the efficacy of TB and MDR-TB treatment and control in Thailand. This project is divided into two phases; the first phase was the evaluation of MDR-TBD-II and INH-RD using M. tuberculosis (Mtb) DNA samples based on multiplex real-time PCR and high-resolution melting curve analysis in TB and MDR-TB diagnosis. The second phase focused on evaluating of the immunomagnetic separation (IMS) using monoclonal antibody against the Mycobacterium Ag85B antigen combined with PCR with confronting two-pair primers (PCR-CTPP), MDR-TBD-II and INH-RD to directly diagnose TB and MDR-TB using clinical sputa. All results will be compared with Xpert MTB/RIF, standard laboratory culture and identification, and DNA sequencing technique. The MDR-TBD-II assay exhibited high performance for TB diagnosis. When it was evaluated with Mtb DNA samples (n=130) against standard culture method, the 100% concordance was achieved. According to the standard proportional method for TB drug susceptibility testing (phenotypic DST) and DNA sequencing method, the specificity of the MDR-TBD-II in identifying RIF-susceptible isolates were as high as 96.7% (58/60) and 98.3% (58/59). In addition, it could detect mutations at codons 531 and 526, the most commonly mutated codons in the rifampin-resistance determining region (RRDR) of rpoB gene in 93.3% (28/30) and 83.3% (10/12). However, this assay had limitations to diagnose the resistance amomg phenotypically RIF-resistant Mtb isolates containing the following genotypes, i) no mutation in the rpoB gene, ii) class-3 SNP or class-4 SNP mutations, and iii) rpoB mutation at codons 516 and 522, which are relatively rare among RIF-susceptible Mtb in Thailand. Diagnosis of isoniazid (INH) -resistant and -susceptible Mtb isolates (n=166) using INH-RD assay showed high concordant results to the phenotypic DST at 98.1% (105/107) and 94.9% (56/59), respectively. This assay could also detect the most commonly found mutations responsible for INH resistance. The katG mutations at codon 315 were detected at 97.9% (95/97) and all mutations of the inhA promoter among tested isolates were detected at 92.3% (12/13). In addition, the results obtained with INH-RD assay were highly in concordance with DNA sequencing method when it was tested with isolates with katG and inhA promoter non-detected mutation (wild type) at 94.9% (56/58). The performance of PCR-CTPP and MDR-TBD-II combined with IMS for direct diagnosis of TB from 259 sputa was evaluated against the culture method. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of PCR-CTPP and MDR-TBD-II were 80.0, 82.4, 84.2, and 77.8%, respectively, and 70.7, 91.6, 90.8, and 72.7%, respectively. Although, the Xpert MTB/RIF and AFB smear conferred a higher sensitivity (97.9% and 93.6%), Xpert MTB/RIF gave a lower specificity (81.5%) than PCR-CTPP และ MDR-TBD-II. However, the Chi-square testing showed that the detection of TB by PCR-CTPP, MDR-TBD-II, Xpert MTB/RIF, and AFB smear is significant different from the culture method (x2=100.11, p-value=0.0064; x2=102.47, p-value=0.0062; x2=170.96, p-value=0.0037; x2=155.92, p-value=0.0041, respectively). In addition, Cochran’s Q test revealed that among IMS-CTPP, MDR-TBD-II, Xpert MTB/RIF, and AFB Smear, there is at least one assay that conferred different results for TB diagnosis (Cochran’s Q=24.32, p-value=0.0000). In comparison to phenotypic DST, the detection of RIF resistance by MDR-TBD-II showed sensitivity and specificity at 70.0% and 69.8%, which were lower than Xpert MTB/RIF (84.85% and 98.57%). The statistical analysis showed that the detection of RIF resistance by MDR-TBD-II is not different from those obtained by phenotypic DST (x2=5.75, p-value=0.11). In contrast, detection of RIF resistance by Xpert MTB/RIF is significantly different from DST results (x2=67.97, p-value=0.0094). Compared to phenotypic DST, the detection of INH resistance by INH-RD showed sensitivity and specificity at 83.3% and 74.2%. The statistical analysis showed that the INH-RD results are not different from the phenotypic DST results (x2=8.94, p-value=0.0709).The results indicated the potential in applying these molecular assays for identification and drug-susceptibility testing of clinical Mtb isolates. However, the application of these methods combined with IMS is limited for the direct diagnosis of TB and drug-resistant TB from clinical specimens. A decreased sensitivity and specificity observed in all methods suggested that the quantity and quality of extracted DNA may be insufficient. Modification of methods for direct extraction of DNA from specimens is, therefore, necessary to reduce the turnaround time (TAT) to as little as 1-2 days. This TAT will be comparable to diagnosis using the Xpert MTB/RIF and TB-LAMP, and faster than the line probe assay. The cost of MDR-TBD-II and INH-RD is inexpensive, approximately 250 bahts per 20-μl reaction (the cost of DNA extraction kit, real-time PCR reagents, and consumables were included). In addition, as high as 96 samples can be tested simultaneously using a 96-well PCR plate. Nevertheless, the asasys require a real-time PCR machine and skilled operators. PCR-CTPP is easy to use, can be used to confirm TB, and is inexpensive (150 baht per 25-μl reaction), but still requires gel electrophoresis to determine PCR results. Therefore, research and development of these methods are essential to build up a prototype technology for TB and MDR-TB diagnosis at the point of care and lead to a sustainable resolution of TB burden in Thailand in the future.
