Detecting latent tuberculosis infection with a breath test using mass spectrometer: A pilot cross-sectional study

Abstract

It is estimated that Mycobacterium tuberculosis (M.tb) infected a quarter of the world's population (1). Latent tuberculosis infection (LTBI) constitutes a broad spectrum of infection states that differ by the degree of pathogen replication, host immune response, and inflammation (2). Approximately 5-10% of those with LTBI will progress to active tuberculosis (ATB) (3). WHO recommends immunodiagnostic tests for LTBI detection, either a tuberculin skin test (TST) or interferon-gamma (IFN-γ) release assays (IGRAs) (4). However, these tests are not precise enough. In certain situations, TB exposure can be used as a surrogate for LTBI (5). Furthermore, TST and IGRAs can not differentiate LTBI from ATB (6). Thus, a more precise tool is urgently needed for the consecutive management of uninfected status, LTBI, and ATB.Recent studies indicate that breathomics may be a useful rule-in or rule-out tool for diagnosing ATB (7), which uncovers the host-pathogen interaction via comprehensive exhaled breath analysis. Breathomics may hold promise to distinguish healthy subjects, LTBI and ATB (8) if a breath test can find the trace and tell the difference of M.tb in consecutive states in the host (9). High-pressure photon ionization time-of-flight mass spectrometry (HPPI-TOFMS) is designed and developed by our team, which can directly detect volatile organic compounds (VOCs) in exhaled breath (10). In our previous studies, this breath detection platform has been verified in lung cancer (11,12), esophagus cancer (13), and Corona Virus Disease 2019 . In this study, we explored the use of this novel, rapid, simple, and inexpensive breath test to detect LTBI.We conducted a cross-sectional study (Chinese Clinical Trials Registry number: ChiCTR2200058346) in which a breath sample was collected from 435 participants with informed consent signed at the Third

Progress and challenges of developing volatile metabolites from exhaled breath as a biomarker platform

“…While a small cohort size, the study provides insight as TB is a leading cause of childhood mortality, and adults and children have different breath profiles. Another study (n = 435) utilized breath analysis to distinguish latent tuberculosis infection (LTBI) from controls, the latter comprising active TB (ATB) and healthy controls (Fu et al, 2023 ). The model showed that breath VOCs could distinguish LTBI from controls with 80.0% sensitivity and 80.8% specificity.…”

Detection of mycobacterial pulmonary diseases via breath analysis in clinical practice

“…In this study, for the first time, we investigated the potential of breath analysis to differentiate among PTB, NTM-PD, OPD and healthy controls (HC), and identified 22 potential VOC biomarkers associated with PTB and NTM-PD, extending our previous PTB breath detection research [17] [18]. Moreover, by employing our self-developed on-line mass spectrometer for human breath-the highpressure photon ionisation time-of-flight mass spectrometry (HPPI-TOF-MS) [20], we built a rapid and simple online breath analysis and modeling method for MPD screening and diagnosis.…”

“…Recently, Beccaria et al used GC×GC-MS to analyse exhaled VOCs of PTB patients and suspected PTB patients / other controls in South Africa / Haiti, achieving high sensitivity and specificity based on 23 / 22 feature VOCs [14] [15], respectively. Other recent studies have focused on pediatric [16] and adult PTB diagnosis [17] and TB infection (TBI) screening [18]. In addition to the above TB VOC-identified studies, sensor-based breath analyses have also shown promise in detecting PTB without biomarkers revealed [7].…”

A multi-omics method for breast cancer diagnosis based on metabolites in exhaled breath, ultrasound imaging, and basic clinical information

“…A standardized workflow has been established using HPPI-TOF-MS for the collection, detection, and data processing of exhaled VOCs, and has validated the feasibility of this technology in detecting diseases such as breast cancer [ 24 ], lung cancer [ 25 ], esophageal cancer [ 26 ], tuberculosis [ 27 ], and Covid-19 [ 28 ]. We aimed to integrate breath-omics, ultrasound radiomics, and clinical features to construct a multi-omics machine learning model, to develop a more accurate, objective, and automated non-invasive diagnostic tool for breast cancer.…”