Yingjie Chen, Qiusheng Yang, Lijuan Chen, Shuo Wei
Forty-five differential metabolites were putatively annotated in the PC versus NC comparison, with 24 remaining significant after FDR correction (q < 0.05). Pathway analysis nominally implicated glycerophospholipid and amino acid metabolism pathways, suggesting that dysregulated lipid metabolism may contribute to the pathogenesis and progression of PC. ROC analysis in this single-center discovery cohort, without independent validation, revealed that alpha-ionone (AUC = 0.8482) and N,N-dimethylsphingosine (AUC = 0.8348) exhibited favorable preliminary discriminatory potential, reflecting distinct metabolic alterations associated with PC. Phenethylamine and phytosphingosine showed moderate preliminary discriminatory potential (AUC = 0.7679 and 0.7567, respectively), and their altered levels raise the possibility of an association with Cryptococcus virulence-related processes. Phytosphingosine was positively correlated with interleukin-6 (IL-6) (r = 0.73, raw P = 0.040), although not after FDR correction (q = 0.423). PC versus PA comparison putatively annotated eight differential metabolites (raw P < 0.05), none remaining significant after FDR correction (q = 0.535); these metabolites were nominally implicated in glycerophospholipid metabolism and polyunsaturated fatty acid-related pathways. Beta-ionone, butaxamine, and cimiracemoside D exhibited gradient trends across the three groups, while cimiracemoside D was positively correlated with erythrocyte sedimentation rate (ESR) (ρ = 0.48, raw P = 0.022), although not after FDR correction (q = 0.423), suggesting metabolic differences between fungal infections may reflect differences in inflammatory responses.
INTRODUCTION: To analyze metabolic alterations associated with pulmonary cryptococcosis (PC) using metabolomic profiling of exhaled breath condensate (EBC) and investigate metabolic differences between fungal infections through comparison with patients with pulmonary aspergillosis (PA).
METHODS: We enrolled 32 patients with PC, 16 patients with PA, and 14 healthy controls (NC) between September 2022 and September 2025. EBC samples were analyzed using ultra-high-performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS). Differential metabolites were identified using t-test, fold-change and PLS-DA. Discriminatory ability was evaluated using receiver operating characteristic (ROC) curves, and potential biological significance was explored through correlation analysis and KEGG pathway.
RESULTS: Forty-five differential metabolites were putatively annotated in the PC versus NC comparison, with 24 remaining significant after FDR correction (q < 0.05). Pathway analysis nominally implicated glycerophospholipid and amino acid metabolism pathways, suggesting that dysregulated lipid metabolism may contribute to the pathogenesis and progression of PC. ROC analysis in this single-center discovery cohort, without independent validation, revealed that alpha-ionone (AUC = 0.8482) and N,N-dimethylsphingosine (AUC = 0.8348) exhibited favorable preliminary discriminatory potential, reflecting distinct metabolic alterations associated with PC. Phenethylamine and phytosphingosine showed moderate preliminary discriminatory potential (AUC = 0.7679 and 0.7567, respectively), and their altered levels raise the possibility of an association with Cryptococcus virulence-related processes. Phytosphingosine was positively correlated with interleukin-6 (IL-6) (r = 0.73, raw P = 0.040), although not after FDR correction (q = 0.423). PC versus PA comparison putatively annotated eight differential metabolites (raw P < 0.05), none remaining significant after FDR correction (q = 0.535); these metabolites were nominally implicated in glycerophospholipid metabolism and polyunsaturated fatty acid-related pathways. Beta-ionone, butaxamine, and cimiracemoside D exhibited gradient trends across the three groups, while cimiracemoside D was positively correlated with erythrocyte sedimentation rate (ESR) (ρ = 0.48, raw P = 0.022), although not after FDR correction (q = 0.423), suggesting metabolic differences between fungal infections may reflect differences in inflammatory responses.
DISCUSSION: EBC metabolomic analysis revealed significant inflammation-related and lipid metabolism-related alterations in patients with PC versus healthy controls, while comparisons with PA suggested that these metabolic changes mainly reflected shared host responses rather than pathogen-specific features.