Breath-test mask detects tuberculosis at the clinic

Tuberculosis (TB) remains the world’s deadliest infectious disease, and current sputum-based tests miss many active cases because people often cannot produce good sputum samples. Breath aerosols, tiny droplets expelled when people breathe and cough, are a major route of Mycobacterium tuberculosis (MTB) transmission and therefore a logical, non-invasive place to look for the bacteria. In a proof-of-principle study led by Tobias Broger, researchers tested a new point-of-care collection kit called the AveloMask that captures breath aerosols on a fiber filter integrated into a clinical-style mask. The study aimed to see whether material collected from the mask could be stabilized, stored, and later tested by PCR to detect active pulmonary TB. By using a mask-based system, the team hoped to simplify sample collection at primary care clinics and reach patients who struggle with sputum collection. The study was carried out with symptomatic adult outpatients in Cape Town, South Africa, and set out to compare mask-based PCR results to standard sputum-based tests and culture methods.

The study enrolled 61 adults with TB symptoms at primary healthcare facilities; 58 provided evaluable breath samples. Participants wore the AveloMask for 45 minutes and were asked to cough deeply five times at the start and end of collection. Breath aerosol was collected on a fiber filter built into the mask, immediately stabilized in buffer after collection, biobanked, and later analysed by quantitative PCR (qPCR) targeting the MTB-specific IS6110 insertion sequence. Diagnostic accuracy was assessed against sputum Xpert MTB/RIF Ultra (SXRS) and a composite microbiological reference standard (MRS) that included culture. Of the 58 evaluable participants, 34 (59%) had confirmed TB. Compared with the SXRS, mask qPCR sensitivity was 71.0% (95% CI: 53.4–83.9%) and specificity 92.3% (95% CI: 75.9–97.9%). Against the MRS the sensitivity was 64.7% (95% CI: 47.9–78.5%) and specificity 91.7% (95% CI: 74.2–97.7%). Mask positivity rose with sputum bacterial load and reached 100% sensitivity in those with high sputum MTB concentrations. Measured IS6110 copy numbers in mask extracts ranged from 4–2147 copies (mean 175), generally low, possibly because of incomplete DNA recovery during lysis or extraction and/or low numbers of MTB bacilli in breath aerosols. Usability questionnaires indicated the mask and procedure were well tolerated.

The findings suggest the AveloMask is a promising and feasible way to sample breath aerosols for molecular detection of active TB at the point of care. Sensitivity and specificity reported here are comparable to other mask-based methods evaluated in research settings, and the device has practical advantages: it is non-invasive, easy to use, and suitable for clinic workflows. The study also highlights technical hurdles to address before broad rollout: improving lysis and extraction to recover more DNA from the collected material, and integrating the collection kit with commercial molecular diagnostic platforms to enable rapid on-site testing. The authors recommend larger and more diverse studies to validate performance in different populations and use-cases, and to determine whether mask-based sampling can reliably identify cases missed by sputum testing. If those steps succeed, AveloMask-style sampling could expand access to TB diagnosis, help detect infectious individuals earlier, and complement existing sputum-based strategies without replacing established reference methods like culture and SXRS.