Development and evaluation of novel bio-safe filter paper-based kits for sputum microscopy and transport to directly detect Mycobacterium tuberculosis and associated drug resistance

Abstract

India has the highest burden of Tuberculosis (TB) and multidrug-resistant TB (MDR-TB) worldwide. Innovative technology is the need of the hour to identify these cases that remain either undiagnosed or inadequately diagnosed due to the unavailability of appropriate tools at primary healthcare settings. We developed and evaluated 3 kits, namely 'TB Detect' (containing BioFM-Filter device), 'TB Concentration and Transport' (containing Trans-Filter device) and 'TB DNA Extraction' kits. These kits enable bio-safe equipment-free concentration of sputum on filters and improved fluorescence microscopy at primary healthcare centres, ambient temperature transport of dried inactivated sputum filters to central laboratories and molecular detection of drug resistance by PCR and DNA sequencing (Mol-DST). In a 2-site evaluation (n = 1190 sputum specimens) on presumptive TB patients, BioFM-Filter smear exhibited a significant increase in positivity of 7% and 4% over ZN smear and LED-FM smear (p<0.05), respectively and an increment in smear grade status (1+ or 2+ to 3+) of 16% over ZN smear and 20% over LED-FM smear. The sensitivity of Mol-DST in presumptive MDR-TB and XDR-TB cases (n = 148) was 90% for Rifampicin (95% confidence interval [CI], 78-96%), 84% for Isoniazid (95% CI, 72-92%), 83% for Fluoroquinolones (95% CI, 66-93%) and 75% for Aminoglycosides (95% CI, 35-97%), using phenotypic DST as the reference standard. Test specificity was 88-93% and concordance was ~89-92% (κ value 0.8-0.9). The patient-friendly kits described here address several of the existing challenges and are designed to provide 'Universal Access' to rapid TB diagnosis, including drug-resistant disease. Their utility was demonstrated by application to sputum at 2 sites in India. Our findings pave the way for larger studies in different point-of-care settings, including high-density urban areas and remote geographical locations.

Fig 1

Developed kits: (A) ‘TB detect’ kit; (B) ‘TB concentration & transport’ kit; and (C) ‘TB DNA extraction’ kit.

Fig 2

(A) Sample processing and staining of BioFM-Filter by using ‘TB detect’ kit; (B) sample processing by using ‘TB concentration & transport’ kit; (C) DNA extraction from Trans-Filter by using ‘TB DNA extraction’ kit.

Fig 3. Study design.

Fig 4. Workflow of the ‘TB detect’ kit evaluation study.

Fig 5. Workflow of the ‘TB concentration & transport’ kit and ‘TB DNA extraction’ kit evaluation study.

Fig 6

(A) Limit of Detection of BioFM-Filter microscopy (40x magnification); (B) PCR amplification of DNA isolated from Trans-Filter. 10, 5, 1 indicate the amount of DNA (in μl) added in PCR and I indicates inhibitor check reaction. Numbers in panels (A) and (B) indicate the number of M. tuberculosis bacteria spiked in 1 ml of sputum; (C) Assessment of stability of DNA on Trans-Filter. Well 0 represents amplification of freshly isolated DNA (day 0) and wells 1 to 4 represent amplification of DNA isolated from Trans-Filters stored at 50ºC at weekly intervals upto 4 weeks. Data for a scanty smear grade sputum sample is shown.

Fig 7. Comparison of smear grade status by ‘BioFM-Filter’ vs. Direct smear microscopy (LED-FM and ZN).

(A) At site 1, NITRD (n = 550); (B) At site 2, TB hospital, Ambala (n = 640); (C) combined performance at both the sites (n = 1190). Left panel: smear grade status; right panel: smear results.

Fig 8. Participant enrolment and testing in ‘TB concentration & transport’ kit and ‘TB DNA extraction’ kit evaluation study.