Tuberculosis (TB) is one of the leading causes of infectious disease worldwide. The diagnosis of TB typically relies on microbiological evidence of the presence of Mycobacterium tuberculosis (MTB) or histological features of the host immune response to MTB in the infected organs. The diagnosis can be enhanced by performing molecular diagnostic tests (e.g. polymerase chain reaction, PCR) on the clinical specimens obtained. Expectorated sputum is usually the first sample sent for MTB culture for suspected pulmonary TB (PTB), which is the most common type of TB. However, this can be particularly challenging for paediatric patients and elderly patients with poor coughing techniques or effort. While for extrapulmonary TB (EPTB), which contributes to 10-20% of TB cases, with TB pleuritis and lymphadenitis as the most common types, invasive investigations are usually required for obtaining clinical specimens of good quality for MTB culture or histological examination. The invasiveness of procedures (e.g. pleural biopsy, lymph node biopsy) and inadequate sensitivity of diagnostic tests could hinder the diagnosis of EPTB. The long turnaround time of MTB culture also creates a challenge for timely diagnosis. Blood sampling for MTB culture or PCR, although non-invasive, has low diagnostic yields. All these urges for non-invasive, rapid and accurate diagnosis of TB.

The standard duration of TB treatment is 6 months, with a longer duration up to 12 months required for certain types of EPTB or in patients with underlying comorbidities (e.g. diabetes mellitus). Treatment monitoring and surveillance for relapses are typically based on a composite of clinical symptoms, sputum MTB culture status, and radiographical appearance. All these domains have their drawbacks, including subjective reporting (clinical symptoms), long turnaround times (sputum MTB culture status), and a lack of diagnostic sensitivity (changes in radiographical appearance in PTB). These clinical unmet needs may be overcome if a non-invasive molecular test could accurately quantify the burden of MTB in the body. Recently, it was reported that the level of MTB cfDNA in plasma can be measured by the CRISPR-TB assay. However, the data were derived mainly from the paediatric patient group and did not evaluate the possibility of latent TB infection (LTBI). This new technology remains explorative at the moment.

Our group has developed a metagenomic sequencing-based assay for measuring the level of MTB cell-free DNA (cfDNA) in plasma. We hypothesize that this new plasma MTB cfDNA assay has the potential to diagnose active TB disease, treatment monitoring and surveillance monitoring by serially measuring the MTB cfDNA level in the plasma. Similar technology may also be applicable to urine, which requires prospective validation.

Inclusion Criteria:

• TB group: patients hospitalized for newly diagnosed TB disease.
• LTBI group: patients without TB disease, but with LTBI diagnosed by either tuberculin skin test (TST) or an interferon-gamma release assay (IGRA) blood test.
• Control group: patients or healthy volunteers without TB disease and LTBI

Exclusion Criteria:

• Initiation of an effective anti-TB treatment regimen more than 24 hours
• Aged 17 years or younger
• Life expectancy of less than 12 months

Patients with active TB disease will be recruited as positive control. Patients with latent TB infection, and subjects without active TB disease and latent TB infection will be recruited as negative control.