Tuberculosis (TB) stages depend greatly on the interaction between () and the host response. As a non-negligible source of active tuberculosis (ATB), individuals with latent infection (LTBI) are insidious and hard to be detected due to limited biomarkers. Further insight into the pathogenic mechanisms associated with heterogeneous clinical outcomes after infection benefits the prevention and control of TB epidemics. Therefore, this study employed four-dimensional data-independent acquisition to quantify the expression level of plasma proteins from healthy controls (HC), LTBI, and ATB, with 15 participants in each group. Key proteins related to TB stages were further assayed using parallel reaction monitoring (PRM) for validation in an independent cohort, with a sample size of 20 individuals per group. Differential abundance analyses showed a notable increase in the number of differentially expressed proteins (DEPs) following infection, mainly involving carbohydrate catabolism, metabolism of cholesterol and lipid, immune response and inflammation, and complement and coagulation cascades. Protein-protein interaction network suggested similar functional enrichment patterns as aforementioned for core DEPs. Weighted gene co-expression network analyses identified six modules, among which the brown module significantly correlated with TB stages. Hub proteins in the brown module were enriched in lipid metabolism and acute-phase response. Furthermore, PRM protein quantification revealed that complement factor H (area under receiver operating characteristic curves [AUC] = 0.708) had a better performance in differentiating LTBI and HC than C4B (AUC = 0.625), while C4B (AUC = 0.917), MBL2 (AUC = 0.887), and SAA1 (AUC = 0.875) were helpful in differentiating ATB and HC. Also, SAA1 (AUC = 0.917) and matrix Gla protein (AUC = 0.905) favored the discrimination of ATB from LTBI. Our work probes into the hub plasma proteins and pathways associated with disease stages following infection, providing critical implications on the diagnosis of TB stages.IMPORTANCEDistinct prognostic outcomes following () infection result from host-pathogen interactions, while the response mechanisms underlying such heterogeneous phenotypes are far from understood. Through four-dimensional data-independent acquisition and parallel reaction monitoring, our study linked specific plasma proteomic profiles to various tuberculosis (TB) stages and corroborated relevant pathways under various disease conditions. Of identified core proteins, complement factor H formed a diagnostic classifier that distinguished latent infection from healthy controls with good performance, and we also identified C4B, MBL2, SAA1, and matrix Gla protein as potential proteomic signatures of active tuberculosis. Additionally, this study further highlighted the critical role of carbohydrate and lipid metabolism, immunological responses, and blood coagulation in TB pathogenesis. Taken together, our findings feature a dynamic landscape of plasma proteome following infection and provide additional evidence on plasma biomarkers for TB diagnosis.