New biomarkers and genes for prediction of type 2 diabetes

Type 2 diabetes is a multifactorial disease caused by non-genetic and genetic risk factors. Current prediction models use mostly non-genetic factors, such as age, overweight, obesity and lifestyle factors, which result in a moderate or good prediction of the individual diabetes risk. This review provides an update on the question to what extent gene variants and metabolites, which are currently the best measurable biomarkers under the new "omics" technologies, can be used to improve risk prediction. Since 2008 microarray-based genome-wide association studies have led to substantially deeper insights into the genetic architecture of type 2 diabetes. This knowledge has improved the understanding of the pathophysiology leading to type 2 diabetes, whereas the predictive value of the novel genetic biomarkers remains fairly low. Metabolomic studies analyze circulating metabolites, such as amino acids and lipids in blood. The predictive value of these metabolites seems to be higher than that of genetic variants. Further large scale studies including whole-genome sequencing will help to obtain a better understanding of the genetic susceptibility for type 2 diabetes. Subsequently, data from different "omics" studies (i.e. genomics, epigenomics, transcriptomics, proteomics and metabolomics) need to be integrated to characterize novel pathogenetic mechanisms and to identify patterns of biomarkers which predict type 2 diabetes better than currently available gene variants and metabolites.