Candidate Obesity Biomarkers Identified Through Multi-Omics Analysis, Mendelian Randomization, and Mediation Analysis

Obesity has become a major public health problem worldwide and a key driver of metabolic disorders. Metabolomics and lipidomics have enabled the discovery of circulating biomarkers, yet their causal relevance remains unclear. In this study, we profiled fasting serum samples from non-obese and obese participants using untargeted ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) metabolomics and liquid chromatography-tandem mass spectrometry (LC-MS/MS) lipidomics to identify obesity-associated metabolites and lipids, and prioritized candidates by integrating two-sample bidirectional Mendelian randomization (MR) and MR-based mediation analyses with immune cell traits based on publicly available genome-wide association studies (GWAS) summary statistics. Receiver operating characteristic (ROC) analyses and regression-based association analyses were further performed to evaluate discriminatory performance and relationships with clinical parameters. We identified hexanoylglycine and lysophosphatidylcholine (LPC (16:0)) as significantly altered in obesity. Both biomarkers showed consistent MR evidence suggesting potential causal relevance to obesity and exhibited associations with obesity status and body mass index (BMI). Two-step MR further suggested that immune cell traits might partially mediate these relationships. In addition, these biomarkers were associated with blood pressure measures, a key indicator of cardiometabolic risk. Collectively, our findings highlight hexanoylglycine and LPC (16:0) as candidate serum biomarkers linked to obesity.