eMedinewS Editorial

Health Care 1,402 Comments

Blood test may predict diabetes risk

Scientists have identified 5 molecules in the blood that can foretell diabetes risk years before typical signs of the disease appear. The finding might help to identify at–risk people who could take steps to delay or halt the disease.

Drs. Thomas J. Wang and Robert E. Gerszten and their colleagues at Massachusetts General Hospital and Boston University analyzed blood samples gathered as part of the Framingham Offspring Study, a long–term community–based study sponsored by NIH’s National Heart, Lung and Blood Institute (NHLBI). The new analysis was funded in part by NIH’s National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). Of more than 2,400 participants with no signs of diabetes in the early 1990s, about 200 went on to develop type 2 diabetes during the 12 years of follow–up. The original blood samples from this latter group were analyzed and compared to samples from age–matched participants who did not develop diabetes but had similar risk factors. The scientists used an approach called metabolomics, which focuses on the unique mix of chemicals, called metabolites that reactions throughout the body leave behind. These chemical fingerprints can provide clues to the body’s health. The researchers used new technologies to rapidly measure the levels of 61 amino acids and other metabolites in the blood samples. The results were reported in the March 20, 2011, advance online edition of Nature Medicine.

The researchers found that elevated levels of 5 amino acids—isoleucine, leucine, valine, tyrosine and phenylalanine—seemed to predict a future diagnosis of diabetes. High levels of these amino acids were detected up to 12 years before onset of disease. Further analysis showed that a combination of 3 amino acids was an even better predictor of diabetes risk. The scientists confirmed their results by analyzing blood samples from more than 300 participants in an independent study of cancer and diet.

Even in participants closely matched for traditional risks factors, such as obesity, these amino acid levels could help differentiate people at greatest risk. Participants with the highest levels of the 3 most predictive amino acids were 4 to 5 times more likely to develop diabetes than those with the lowest levels.
Dr KK Aggarwal
Editor in Chief