A tuberculosis test that works with nothing more than a tongue swab could transform how the world’s deadliest infectious disease is detected, according to new research from Tulane University that addresses a critical gap in global health screening.
The enhanced CRISPR-based test showed 74% accuracy in detecting TB from tongue swabs compared to just 56% for traditional methods, potentially offering a painless alternative to the sputum-based testing that has dominated TB diagnosis for decades. With an estimated 4 million tuberculosis cases going undiagnosed annually, largely due to testing difficulties, this development represents a significant step toward more accessible community-based screening.
“More than 10 million people worldwide fall ill with tuberculosis every year, but 40% of those cases are considered missing as people go undiagnosed. To find those missing cases, testing needs to be less invasive and more accessible to reach as many people as possible who may not otherwise be tested.”
The research, published in Nature Communications, tackles a longstanding problem in TB diagnosis. Current testing relies on collecting sputum from the lungs and lower respiratory system, a process that proves impossible for about 25% of symptomatic patients and nearly 90% of asymptomatic cases. Children, HIV patients, and those with certain forms of TB often cannot produce the thick mucus required for traditional testing.
Beyond the Lab, Into Communities
The new method, called ActCRISPR-TB, can return results in under an hour using a “one pot” approach similar to COVID-19 rapid tests. A swabbed sample goes into a pre-loaded tube with a test strip and reagent, gets incubated for 45 minutes, then displays colored bands indicating infection status.
Clinical testing across 603 patient samples demonstrated the method’s versatility beyond tongue swabs. The test achieved 93% sensitivity with adult respiratory samples, 83% with pediatric stool samples, and 93% with adult spinal fluid specimens. This range of sample types could prove crucial for populations who struggle with traditional sputum-based testing.
The research team, led by corresponding author Tony Hu and lead author Zhen Huang, tested their approach across diverse patient populations including HIV-positive adults and children suspected of having TB. In one particularly challenging group, HIV-positive patients with severely compromised immune systems, the new test detected 80% of TB cases while the standard Xpert test caught only 40%.
Point-of-Care Potential
Perhaps most significantly for global health applications, the researchers developed a lateral flow version that works without laboratory equipment or trained medical staff. This smartphone-readable format maintained comparable accuracy while requiring only simple thermal processing of samples.
“Tongue swabs are painless, easy to collect, and don’t require trained medical staff. That opens the door to large-scale screenings.”
The tongue-based approach builds on recent understanding that TB bacteria expelled from infected lungs accumulate on the tongue surface, providing direct evidence of disease. However, the bacterial concentrations are much lower than in sputum, requiring the enhanced sensitivity that CRISPR technology provides.
Testing revealed some practical considerations for implementation. Morning tongue swabs showed higher detection rates than evening samples, and proper sample preservation proved important for maintaining accuracy. Food residue on the tongue affected signal strength but not the ability to identify positive cases.
The study represents the latest advancement in Hu’s broader effort to move TB testing out of laboratories and into communities where the disease spreads. His team has previously developed rapid tests that work on smartphone-sized devices and others requiring no electricity, while also using artificial intelligence to identify drug-resistant strains.
While acknowledging that further research and development remains necessary, the authors see this work as significant progress toward ending TB as a public health threat. The approach addresses multiple barriers to effective TB screening: complex sample collection, laboratory requirements, and limited accessibility in resource-constrained settings where the disease burden remains highest.
The technology could prove particularly valuable for active case-finding efforts in communities with high TB transmission rates, where the current diagnostic gap contributes to ongoing spread of the disease. With tuberculosis remaining a leading infectious disease killer globally, tools that can identify cases earlier and more easily represent critical advances in the fight against this ancient scourge.
Nature Communications: 10.1038/s41467-025-63094-x
ScienceBlog.com has no paywalls, no sponsored content, and no agenda beyond getting the science right. Every story here is written to inform, not to impress an advertiser or push a point of view.
Good science journalism takes time — reading the papers, checking the claims, finding researchers who can put findings in context. We do that work because we think it matters.
If you find this site useful, consider supporting it with a donation. Even a few dollars a month helps keep the coverage independent and free for everyone.