Objective: We validated the accuracy of host selected signature gene set using unstimulated whole blood (WB), and peripheral blood mononuclear cells (PBMC) in the diagnosis of tuberculosis (TB).
Methods: The unstimulated WB and PBMC from 1417 individuals with active pulmonary TB patients, other lung diseases and healthy participants were analyzed using real time polymerase chain reaction (RT-PCR).
Results: The WB cohort test demonstrates that the combination of GBP5 and KLF2 can differentiate active TB versus HC with sensitivity and specificity of 77.8% and 87.1%, respectively; but most importantly active TB versus OD with sensitivity and specificity of 96.1% and 85.2%, respectively. Again during treatment course, the TB score of GBP5 and KLF2, analytes secretion and clinical parameters were found to be associated in disease progression. In the PBMC cohort test, we found that the only and best discriminatory combination was GBP5, DUSP3 and KLF2 inthe active TB versus HC with a sensitivity and specificity of 76.4% and 85.9%, respectively.
Conclusions: Our study reveals that GBP5 and KLF2 may be useful as a diagnostic tool for active TB, also the two-gene set may serve as surrogate biomarkers for monitoring TB therapy.
Keywords: Diagnosis; Mycobacterium tuberculosis; Other diseases; Signature gene.
Copyright © 2017. Published by Elsevier Ltd.