Diabetic nephropathy is increasing worldwide however no new treatments have been approved for DN in the past 20 years(1). Despite the identification of many new targets and development of new therapeutics, a major obstacle remains adequate biomarkers to gauge likelihood of benefit of new drugs in the development phase. We have taken a urinary-based metabolomic approach to identify biomarkers for DN. Using a targeted metabolomics strategy we identified that a panel of metabolites are characteristic of patients with diabetes and reduced renal function (eGFR <60 ml/min/1.73 m2)(2). These metabolites comprises a large network of interacting pathways and largely indicate a reduction in mitochondrial function. A reduction of mitochondrial function was demonstrated in kidney biopsy tissues and in urine exosomes. Furthermore, the major underlying pathway leading to reduced mitochondrial function appears to be the AMPK-PGC1a pathway (3). The metabolomic signature may prove to be a non-invasive marker of underlying mitochondrial function and could be used to monitor disease progression and possibly responses to novel treatments. Several novel therapeutic agents provide renoprotection (4) and are able to stimulate mitochondrial biogenesis and increase urinary levels of the metabolomic panel of diabetic kidney disease. These studies suggest that metabolomics is a powerful platform to identify new therapeutic targets and to monitor drug development for diabetic complications.