Stool sequencing finds new TB targets but not better detection

Finding active tuberculosis (TB) in young children is difficult because they often cannot produce sputum for testing. Stool-based molecular tests are a noninvasive alternative, but they typically show lower sensitivity than sputum-based tests. To explore whether a broader, untargeted approach could help, a research team led by Eric Wobudeya evaluated metagenomic sequencing of stool DNA as a diagnostic tool. The study analyzed 176 stool samples from children with a median age of 3.6 years (IQR, 1-6 years) who were being evaluated for possible TB. The idea behind metagenomic sequencing (mNGS) is to search broadly for Mycobacterium tuberculosis genetic material and related bacterial signals in stool without restricting the search to a few predefined targets, with the hope of improving detection and discovering new gene or protein targets that could be used in future molecular tests. The research was carried out in a real-world pediatric setting in Uganda and directly compared stool mNGS results with established stool molecular testing approaches.

The team used untargeted metagenomic sequencing (mNGS) on stool DNA and classified sequencing reads to detect Mycobacterium tuberculosis signals. Detection rates using thresholds of ≥1, ≥2, or ≥5 sequence fragments were reported as 35.5%, 45%, and 19.4% respectively, with specificities of 92.64% (87%-96%), 97% (93%-99%), and 99.3% (96%-100%). When all samples were considered, stool Xpert Ultra had a similar sensitivity (22.6%) to stool mNGS. In a head-to-head comparison, stool mNGS had lower sensitivity than stool Xpert Ultra (38.5% vs. 53.8%, difference -15.3%, 95% CI 14-68 to 25-81). Beyond simple detection, metagenomic sequencing identified genomic and protein signals—specifically rRNA, virulence proteins and membrane proteins—that are not targeted in current PCR-based platforms. The study therefore measured both diagnostic performance and the potential of mNGS to uncover new molecular markers.

The main conclusion of the work led by Eric Wobudeya is clear: metagenomic sequencing of stool DNA did not increase the sensitivity of TB detection compared with existing stool tests, but it did reveal novel targets that current PCR tests do not cover. That means mNGS may not be ready to replace stool Xpert Ultra as a diagnostic tool based on sensitivity alone, but it can play an important discovery role. The rRNA, virulence proteins and membrane proteins detected by mNGS could inform the development of next-generation molecular assays that are more sensitive when applied to stool samples. For clinicians and test developers, the study suggests a two-part strategy: continue to use the best current stool diagnostics for patient care while using metagenomic discovery efforts to expand the menu of targets that future tests might include. In short, mNGS may be less useful as a frontline diagnostic today but valuable as a source of new diagnostic markers for tomorrow.