Discovering novel biomarkers for diagnosis and treatment monitoring of active pulmonary tuberculosis by ion metabolism analysis

Abstract

Tuberculosis (TB) is a highly lethal infectious disease that poses a global threat. Timely and accurate biomarker for TB diagnosis and treatment monitoring remains a pressing need. Ions, the crucial trace element for humans, may be potential targets for TB diagnosis and the forecasting of TB development. To explore the potential of ions as biomarkers, we measured and compared the levels of various ions in whole blood and plasma samples from healthy control (HC), pulmonary TB patients (TB), cured pulmonary TB patients (RxTB), and other non-TB pneumonia patients (PN) by using ultra-high performance liquid chromatography-tandem mass spectrometry. Our study demonstrated that Cu (AUC = 0.670), Pb (AUC = 0.660), and Zn (AUC = 0.701) in whole blood exhibited promising diagnostic performance for TB. Then we used a neural network (NNET) for TB prediction, the AUC values used to differentiate definite TB from HC or PN in plasma were 0.867 and 0.864, respectively. The AUC values used to differentiate definite TB from HC or PN in whole blood were 0.818 and 0.660, respectively. Our correlation analysis showed that Zn (r= 0.356, p=0.001) and Cu (r= 0.361, p=0.0004) in plasma are most closely related to disease severity. Additionally, six ions (Cu, Sb, V, Mn, Fe, Sr) in plasma and whole blood were altered following anti-TB therapy. These results showed that ions could be diagnostic biomarkers for TB. Furthermore, the level of particular ions can forecast the degree of lung damage and the success of the TB treatment. In conclusion, this study highlights the possibility of using ions from blood samples to enable rapid tuberculosis diagnosis.