A novel Three-miRNA diagnostic signature for tuberculosis: Non-invasive detection and stratification from latent to active disease.

BACKGROUND

Tuberculosis (TB) is the leading infectious killer and about a quarter of the world's population is estimated to have latent TB infection (LTBI). Present diagnostics do not convincingly separate latent from active TB cases; thus, there is an urgent need for new biomarkers to facilitate accurate diagnosis and risk stratification.

OBJECTIVES

This study seeks the development and validation of a three-microRNA signature consisting of miR-29a, miR-144, and miR-199a in the diagnosis and risk stratification of tuberculosis (TB), gradients of expression across the spectrum be established cut-off values that would work optimally to assist clinical decision-making.

MATERIALS AND

METHODS

A cross-sectional study was carried out on 100 participants comprising 30 healthy controls (HC), 35 with latent tuberculosis infection (LTBI), and 35 patients suffering from active TB. Quantification of microRNA expression levels was done using real-time polymerase chain reaction (PCR). The diagnostic accuracy of this microRNA signature was determined through ROC curve analysis. The applied statistical methods were ANOVA, post hoc testing, and correlation analysis.

RESULTS

The three-microRNA signature showed graded expression across the spectrum of TB. MiR-29a revealed maximum differences in expression levels between groups (HC: 1.200 ± 0.485, LTBI: 2.644 ± 1.729, Active TB: 7.380 ± 6.138; p < 0.001). MiR-144 and miR-199a have followed a similar pattern of expression. The combined microRNA signature has excellent diagnostic accuracy with an area under the curve (AUC) of 0.92 for the separation of active TB from LTBI, and good performance yielding an AUC of 0.84 when distinguishing LTBI from healthy control.

CONCLUSION

The three-microRNA signature represents a promising non-invasive blood-based approach for differentiating healthy controls, LTBI, and active TB in this single-center cross-sectional study. The observed expression gradients and diagnostic performance support further evaluation in larger, independent, multi-center cohorts.