Comparative assessment of host transcriptomic- and bacteriological- markers for monitoring treatment response among active tuberculosis patients: a longitudinal study.

OBJECTIVES

Accurate diagnosis and effective treatment response monitoring (TRM) remain major challenges in tuberculosis (TB) control. Most existing tools are sputum based, yet many patients are unable to produce sputum at diagnosis or during treatment. We evaluated blood-based host transcriptional markers (HTM) for active TB diagnosis and TRM.

METHODS

Presumptive TB patients presenting to healthcare centres in Blantyre, Malawi and community healthy controls were enrolled. Active TB (ATB) was confirmed using Xpert MTB/RIF and Mycobacterial Growth Indicator Tube (MGIT) culture. Participants negative by both MGIT and Xpert were classified as having other respiratory diseases (ORDs). Confirmed ATB cases initiating treatment were followed for 8 months with serial sputum bacteriology (MGIT time to positivity (TTP) and TB Molecular Bacterial Load Assay (TB-MBLA) and HTM measurement. HTM expression was quantified in whole blood using reverse-transcriptase quantitative PCR.

RESULTS

Before treatment, CD64, BATF2, GBP5 and C1QB best discriminated ATB from ORDs, with areas under the curve of 96.3%, 94.4%, 93.9% and 93.7%, respectively. All participants responded clinically by 2 months of therapy. Median MGIT TTP increased from 7 days (IQR 5-12) at baseline to 42 days, while mean bacterial load declined from 3.87 (SD 1.4) to 1.4 (SD 0.5) log10 eCFU/mL by month 8. HTM expression declined in a pattern consistent with bacteriological markers and normalised to levels observed in TB-free individuals by end of follow-up, irrespective of HIV status (P < 0.0001).

CONCLUSION

Blood-based HTM closely track bacteriological response and offer a promising sputum-independent adjunctive approach for diagnosis and treatment monitoring of pulmonary TB.