Discovery of biomarkers for enhanced prediction of type 2 diabetes outcome: Type 2 diabetes is a highly heterogenic disease that can involve multiple mechanisms deficient to a varying degree ultimately leading to hyperglycemia, this make administering a silver bullet treatment difficult, as response can be individual.
We use a machine learning approach to incorporate standard clinical measurements and metabolomic profiles from blood samples, to better predict individual treatment outcome and evaluation complication risk.
The aim is to assess the risk of progression of pre-diabetes to overt diabetes and of rapid deterioration of type 2 diabetes.
This is a part of a collaboration of 20 academic research laboratories focused especially on diabetes research and two Small and medium-sized enterprises (one specialized in lipid analyses and the other in scientific project management) and four well-established pharmaceutical and biotechnology establishments, all part of the European Federation of Pharmaceutical Industries and Associations (EFPIA).
The purpose of POPs is to find additional, preferably biological, measurements that can help to stratify T2D patients into smaller more homogeneous subgroups using metabolomics, in order to enhance the efficacy of treatments given to each individual patient