Purpose: Tuberculosis (TB) is the leading cause of mortality in individuals infected with human immunodeficiency virus (HIV), yet the methods for detecting Mycobacterium tuberculosis  at an early stage remain insensitive or ineffective. This study aimed to discover plasma biomarkers for distinguishing HIV-TB coinfected individuals from HIV individuals without TB (HIV-nonTB).
Patients and Methods: A total of 200 Chinese HIV-positive patients were recruited, 100 each for HIV-nonTB group and HIV-TB group. Plasma proteomic profiles were analyzed for 50 patients each in both groups, using data-independent acquisition (DIA)-mass spectrometry-based proteomics. Differently expressed proteins were revealed with ridge regression analysis. Enzyme-linked immunosorbent assay (ELISA) analyses were performed for further validation in other 100 patients.
Results: DIA-mass spectrometry revealed 13 upregulated and 33 downregulated proteins in the HIV-TB group. AMACR (α-methylacyl-CoA racemase), LDHB (L-lactate dehydrogenase B chain), and RAP1B (Ras-related protein Rap-1b) were selected for building a diagnostic model, for which the receiver operation characteristic curve had under areas of 0.99 and 0.89 testing with proteomics data (sensitivity = 92%, specificity = 100%) and ELISA data (sensitivity = 76%, specificity = 92%), respectively.
Conclusion: The combination of AMACR, LDHB, and RAP1B proteins may serve as a potential marker of TB in HIV-infected patients.
Keywords: diagnosis, coinfection, AIDS-related opportunistic infections, Mycobacterium tuberculosis, ROC curve, proteome