RESUMO
Background: The currently used conventional susceptibility testing for drug-resistant Mycobacterium tuberculosis (M.TB) is limited due to being time-consuming and having low efficiency. Herein, we propose the use of a microfluidic-based method to rapidly detect drug-resistant gene mutations using Kompetitive Allele-Specific PCR (KASP). Methods: A total of 300 clinical samples were collected, and DNA extraction was performed using the "isoChip®" Mycobacterium detection kit. Phenotypic susceptibility testing and Sanger sequencing were performed to sequence the PCR products. Allele-specific primers targeting 37 gene mutation sites were designed, and a microfluidic chip (KASP) was constructed using 112 reaction chambers to simultaneously detect multiple mutations. Chip validation was performed using clinical samples. Results: Phenotypic susceptibility of clinical isolates revealed 38 rifampicin (RIF)-resistant, 64 isoniazid (INH)-resistant, 48 streptomycin (SM)-resistant and 23 ethambutol (EMB)-resistant strains, as well as 33 multi-drug-resistant TB (MDR-TB) strains and 20 strains fully resistant to all four drugs. Optimization of the chip-based detection system for drug resistance detection showed satisfactory specificity and maximum fluorescence at a DNA concentration of 1×101 copies/µL. Further analysis revealed that 76.32% of the RIF-resistant strains harbored rpoB gene mutations (sensitivity, 76.32%; specificity 100%), 60.93% of the INH-resistant strains had katG gene mutations (sensitivity, 60.93%; specificity, 100%), 66.66% of the SM-resistant strains carried drug resistance gene mutations (sensitivity, 66.66%; specificity, 99.2%), and 69.56% of the EMB-resistant strains had embB gene mutations (sensitivity, 69.56%; specificity, 100%). Further, the overall agreement between the microfluidic chip and Sanger sequencing was satisfactory, with a turnaround time of the microfluidic chip was approximately 2 hours, much shorter than the conventional DST method. Conclusion: The proposed microfluidic-based KASP assay provides a cost-effective and convenient method for detecting mutations associated with drug resistance in M. tuberculosis. It represents a promising alternative to the traditional DST method, with satisfactory sensitivity and specificity and a much shorter turnaround time.
RESUMO
BACKGROUND: Sputum is commonly used for the diagnostic testing of pulmonary tuberculosis (PTB), but people living with HIV/AIDS (PLWHA) usually have little sputum. Moreover, the automated molecular test, Xpert MTB/RIF assay (Xpert), has a low sensitivity in PLWHA. We aimed to estimate the performance of Xpert Ultra on the detection of Mycobacterium tuberculosis (MTB) using bronchoalveolar lavage (BAL). METHODS: From February 5, 2018 to March 30, 2019, a total of 99 PLWHA with suspected PTB at the Third People's Hospital of Shenzhen, China, were recruited. The information on demographics and medical history, blood MTB antigen-specific interferon gamma enzyme-linked immunospot assay (T-SPOT.TB), T lymphocyte subsets, and plasma HIV RNA load were collected. Computed tomography (CT) and flexible bronchoscopy were performed, and BAL and blood samples were collected. Testing of acid-fast bacilli (AFB), tuberculosis real-time fluorescence quantitative PCR (TBDNA), Ultra, Xpert, and MTB culture were conducted. RESULTS: Compared to BAL MTB culture for tuberculosis diagnosis, Ultra, Xpert, T-SPOT.TB, TBDNA and AFB smear had the sensitivity of 0.96 (24/25), 0.80 (20/25), 0.84 (21/25), 0.44 (11/25), and 0.12 (3/25), respectively; and the specificity of 0.92 (68/74), 0.96 (71/74), 0.93 (69/74), 0.96 (71/74), and 0.99 (73/74), respectively. Our study found that the sensitivity of Ultra was higher than that of culture and Xpert (AUC 0.92, 0.86 and 0.84, respectively). The results also indicated that PLWHA with CD4 <200 cells/mm3 had reduced both sensitivity (from 1.00 and 0.86 to 0.94 and 0.78, respectively) and specificity (from 0.96 and 1.00 to 0.90 and 0.41, respectively) of Ultra and Xpert for the diagnosis of PTB. DISCUSSION: Our data supported an increased sensitivity of Ultra compared to that of Xpert on BAL samples of PLWHA, regardless of the CD4 counts and reference diagnosis standards.
