• Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, MD, USA.
• Division of Epidemiology, Department of Medicine, Vanderbilt Epidemiology Center, Institute for Medicine and Public Health, Vanderbilt-Ingram Cancer Center, Vanderbilt University School of Medicine, Nashville, TN.
• Shanghai Cancer Institute, Shanghai, China.
• University of Texas MD Anderson Cancer Center, Houston, TX, USA.
• Department of Epidemiology and Preventive Medicine, Regensburg University Medical Center, Franz-Josef-Strauss-Allee 11, 93053 Regensburg, Germany.

BACKGROUND: A high body mass index (BMI) is a major risk factor for several chronic diseases, but the biology underlying these associations is not well-understood. Dyslipidemia, inflammation, and elevated levels of growth factors and sex steroid hormones explain some of the increased disease risk, but other metabolic factors not yet identified may also play a role.

DESIGN: In order to discover novel metabolic biomarkers of BMI, we used non-targeted metabolomics to assay 317 metabolites in blood samples from 947 participants and examined the cross-sectional associations between metabolite levels and BMI. Participants were from three studies in the United States and China. Height, weight, and potential confounders were ascertained by questionnaire (US studies) or direct measurement (Chinese study). Metabolite levels were measured using liquid-phase chromatography and gas chromatography coupled with mass spectrometry. We evaluated study-specific associations using linear regression, adjusted for age, gender, and smoking, and we estimated combined associations using random effects meta-analysis.

RESULTS: The meta-analysis revealed 37 metabolites significantly associated with BMI, including 19 lipids, 12 amino acids, and 6 others, at the Bonferroni significance threshold (p<0.00016). Eighteen of these associations had not been previously reported, including histidine, an amino acid neurotransmitter, and butyrylcarnitine, a lipid marker of whole-body fatty acid oxidation. Heterogeneity by study was minimal (all P_{heterogeneity} >0.05). In total, 110 metabolites were associated with BMI at the p<0.05 level.

CONCLUSION: These findings establish a baseline for the BMI metabolome, and suggest new targets for researchers attempting to clarify mechanistic links between high BMIs and disease risk.