A multi-omics case-control study identifying oropharyngeal microbiome-metabolite patterns that characterize secondary bacterial pneumonia among influenza patients.

Secondary bacterial pneumonia is a severe complication of influenza;howeve the biological determinants that distinguish progression from uncomplicated infection remain poorly understood. We investigated the oropharyngeal microbiome and plasma metabolome as potential discriminators of pneumonia development. In this study, we report a cross-sectional case-control study conducted during the 2022-2023 influenza season to identify and internally validate a microbiome-metabolite profile that characterizes pneumonia cases from uncomplicated influenza. We enrolled 236 consecutive influenza patients from Jiangsu Province, China (October 2023-December 2024): 59 with secondary pneumonia and 177 uncomplicated controls. Oropharyngeal swabs were subjected to 16S rRNA V3-V4 sequencing; plasma metabolomics was performed by UPLC-MS/MS in both ion modes. Seven machine-learning algorithms were compared; Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression was selected because it yielded the highest cross-validated discrimination. Microbial composition distinguished groups, not richness. Pneumonia cases showed enrichment of Synergistota and Bifidobacteriaceae with depletion of Bacillaceae (β-diversity p = 0.057). Controls exhibited enriched glycolysis and lipid metabolism pathways; pneumonia cases showed elevated degradation pathways (GLUCARDEG and GALLATE-DEGRADATION). Plasma metabolomics revealed a lipid depletion signature: phospholipids PC(O-16:0/0:0) and PS(14:0/18:3(9Z,12Z,15Z)) were significantly reduced (area under the (receiver operating characteristic) curves (AUCs) = 0.69-0.71). Small Molecule Pathway Database (SMPDB) pathway analysis demonstrated suppressed anabolic (tyrosine, steroid, and purine metabolism) and enhanced catabolic (beta-oxidation of very long-chain fatty acids) pathways. Machine learning identified Peptococcus as the top indicator (LASSO AUC = 0.65); Shapley Additive Explanation (SHAP) analysis revealed a monotonic risk increase with abundance. Oropharyngeal dysbiosis and systemic metabolic reprogramming characterize influenza cases that progress to secondary pneumonia. Peptococcus and four metabolites form an internally validated exploratory profile associated with secondary pneumonia; external validation and performance optimization are warranted.