Back

Plasma multiomics distinguishes pulmonary tuberculosis from other respiratory infections

Mousavian, Z.; Nabeemeeah, F.; Nellis, M. M.; Gandhi, N. R.; Kempker, R. R.; Jones, D. P.; Johnson, H.; Islam, M.; Magee, M. J.; Martinson, N.; Sharma, A. A.; Collins, J. M.

2026-03-19 infectious diseases
10.64898/2026.03.18.26348671 medRxiv
Show abstract

Novel blood-based biomarkers for tuberculosis (TB) are needed to develop rapid, point-of-care diagnostics. We sought to use combined plasma metabolomics and high-density cytokine profiling to identify a biomarker signature that can differentiate pulmonary TB (PTB) from patients hospitalized with other respiratory diseases and ambulatory household contacts with TB symptoms. We analyzed plasma concentrations of 28 cytokines and 118 metabolites from 391 adults ([≥]18 years) presenting with respiratory symptoms suggestive of TB, of which 187 had PTB confirmed by Xpert MTB and/or M. tuberculosis sputum culture and 204 were controls in whom PTB was excluded. Our study identified a 5-marker signature (IFN.gamma, IL.22, IL.10, methionine and oxoproline) with an AUC of 0.97 (95% CI: 0.95-1.00) in the test set. The signature had 98% and 84% sensitivity at 70% and 98% specificity respectively, which meet WHO target product profiles for both non-sputum triage and diagnostic TB tests.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.