Obesity is a significant public health burden with nearly 40% of the world’s population classified as overweight and 13% as obese. Body mass index(BMI) results from a complex interaction between lifestyle, environmental factors and underlying genetic susceptibility. The metabolome represents a dynamic functional readout of the state of a biological system; encompassing both genetic and environmental influences. Consequently, metabolomics is ideally suited to explore the drivers and manifestations of BMI on a mechanistic and metabolic level. We are conducting a meta-analysis within the context of the global COnsortium of METabolomics Studies (COMETS). Participating COMETS cohorts with plasma metabolomic profiling and concurrent measures of BMI. We utilized models that explore the correlation between BMI and metabolite levels, with adjustment for and stratification by potential effect modifiers. These models are run independently within all participating cohorts and then the results are meta-analyzed using random and fixed effects models. Bonferroni is used to adjust for multiple testing and the Cochran’s Q test and the Wald test are used to explore between-study and between-strata heterogeneity respectively. A total of 43 cohorts encompassing >98,000 participants have enrolled in the study to date. Preliminary findings using 8 cohorts and 4450 subjects, there were 256 metabolites that could be harmonized across all cohorts. In a baseline model adjusting for gender, age, and race, 137(53.5%) metabolites were significantly associated with BMI after Bonferroni correction. The top hit was the amino acid Glutamtate (meta-analyzed spearman’s r=0.33(1.1x10-47). A number of other metabolites demonstrated significant heterogeneity across the cohorts that was primarily driven by gender and fasting status. An updated analysis of ~30,000 participants retained these initial findings. These analyses confirm the feasibility of large-scale meta-analyses of metabolomics and suggest an important role for glutamate metabolism in BMI, which may be acting through hypothalamic regulation of appetite, insulin sensitivity or dyslipidemia. The results also suggest the metabolome of BMI may differ between genders and that BMI-plasma metabolite relationships are sensitive to fasting status at the time of blood draw.