"We have identified a set of urine proteins almost 10 years before the onset of diabetic nephropathy that identify those who will go on to get diabetic nephropathy, the most common cause of renal failure in the U.S. and the world," Dr. Ravi Thadhani, from Massachusetts General Hospital and Harvard Medical School in Boston, told Reuters Health.

Dr. Thadhani and colleagues compared urinary proteomic profiles among 62 Pima Indians with type 2 diabetes and normal urinary albumin excretion, who were followed for 10 years for the development of diabetic nephropathy.

The authors identified a 12-peak predictive signature that provided a 93% sensitivity, 86% specificity, and 93% accuracy for predicting the development of diabetic nephropathy.

When the signature was tested against all samples in the validation set (17 case samples and 17 matched controls), the overall accuracy was 74%, with a sensitivity of 71% and a specificity of 76%, the report indicates.

The 12-peak signature was independently predictive of diabetic nephropathy in a multivariate binary logistic regression model adjusting for baseline hemoglobin A1c, the researchers note, whereas hemoglobin A1c was no longer significantly associated with subsequent diabetic nephropathy.

"While these findings require a significant amount of work to identify and test the robustness of these markers, they offer the potential for clinicians one day to be able to tell whether a diabetic patient will get renal failure well before the development of this devastating condition, and, hence, offer hope for early intervention to prevent its onset," Dr. Thadhani said.

"First we need to confirm the identification of the proteins we uncovered, then we need to test this profile in other populations with type 2 diabetes to determine if the same predictive potential remains," Dr. Thadhani explained. "Then we will carry out prospective studies in larger populations."

Diabetes Care 2007;30:638-643.

=======================================

DID YOU KNOW: