Circulating hexose-derived metabolites are biomarker candidates for diabetes risk prediction

Abstract

Abstract Importance Diabetes mellitus type 2 (DMT2) is an important risk factor for the development of cardiovascular disease (CVD). While risk stratification for CVD itself remains difficult given the complex causative mechanism of the disease, predicting DMT2 is even more challenging. Metabolomic profiling offers the potential to detect new biomarkers and improve risk assessment of CVD and potentially DMT2. Objective and Methods To evaluate the association between circulating metabolites and incident diabetes in a large European population cohort. This study utilized the Biomarkers for Cardiovascular Risk Assessment in Europe (BiomarCaRE) case-cohort to measure circulating metabolites using a targeted approach in serum samples from 9,847 individuals without a history DMT2. The cohort consisted of a weighted, random subcohort of the original cohort of more than 70,000 individuals. The case-cohort design was applied to 6 European cohorts. Associations with time to DMT2 onset were assessed individually. The association of metabolites with incident DMT2 was examined using Cox regressions and C statistics. Results In 9,847 individuals (3,989 women (40.5%); median (interquartile range) age, 56.0 (48.8, 61.6), 1733 incident DMT2 cases (17.6%) occurred over a median (interquartile range) follow-up time of 8.0 (4.3, 14.3) years. Among the 141 metabolites analyzed, 59 remained significantly associated with incident DMT2 after correction for multiple testing. The metabolite H1 showed a much lower p-value (1.9x10-110) in Cox regression analyses than all other metabolites. It also returned a hazard ratio (HR) of 6.88 (5.81, 8.15) and a HR per standard deviation (SD) of 1.61 (1.55, 1.68). This was the case for both women: HR per SD 1.56 (1.46; 1.66), p=3.48x10-40 and men: 1.65 (1.56; 1.74), p=5.25x1073 and was consistent across all sub-cohorts. H1 describes a signature of metabolites derived from hexoses with the molecule formula C6H12O6, such as glucose. Adding H1 to a base risk model including classic risk factors, high-sensitivity C-reactive protein and high-sensitivity troponin I improved discrimination by C statistics significantly: C statistics: H1 alone 0.825 (0.800, 0.850), base model + H1 0.858 (0.833, 0.883); base risk model alone: 0.810 (0.785, 0.835); p(model + H1 vs. model alone)<0.001. At the same time, H1’s correlations with classic CVD risk factors was low. Conclusions and Relevance In a large population cohort of >8,000 individuals we tested association of metabolites with incident DMT2. H1 significantly improved prediction of DMT2 when added to a classic risk factor model and represents a promising signature of circulating biomolecules for DMT2 risk prediction. Additionally, 59 metabolites were significantly associated with risk of incident DMT2. Furthermore, these findings hold promise for an improved understanding of the pathophysiology in the development of metabolic syndrome and DMT2 in its early stage.