Will ADA endorse new measure for glycemic control?

October 8, 2007

Changes are being evaluated in the value used to assess long-term glycemic control-moving from HbA1cto average blood glucose.

At a presentation held recently at ADA's 68th Scientific Sessions in Chicago, David M. Nathan, M.D., said he believes there is an improved measure that will enable patients to better understand their disease. Nathan, a professor at Harvard Medical School and chief of the diabetes unit medical service and the Diabetes Center at Massachusetts General Hospital, said that average blood glucose, which could be reported in the same units that patients are familiar with, would be easier for them to understand.

"It would have the advantage of reporting chronic glycemia in the same units as the patient's self-monitoring," Nathan said during the presentation. Specifically, the value would be reported as A1c-derived average glucose (ADAG), in mg/dl, instead of the currently reported A1c percentage. The question, however, is whether an accurate and reliable regression equation can be developed that converts A1c into ADAG. To that end, Nathan is waiting for the final results of his clinical trial designed to answer just that question.

At the ADA session, Nathan presented the preliminary results of 254 patients from nine of the 10 clinical centers. He warned the audience that these were only data completed as of June 1 and represented only an initial analysis. "Things could change before the European Association for the Study of Diabetes [EASD]" meeting," he said, where he planned to present his final results last month.

Three models examined

Of the three models examined, a linear regression and an exponential model showed promise. The linear correlation of average glucose over three months with an A1c at month three showed a highly significant correlation. "We'll figure out which model is the best fit," Nathan said. "One question is, How well are we going to determine if the relationship is durable enough to have us feel confident?" In response to his own inquiry, Nathan said he and his researchers set a 90% tolerance interval-where 90% of subjects need to fit within plus or minus 15% across the entire range of average glucose.

In agreement with Nathan is Richard Kahn, Ph.D., chief scientific and medical officer for ADA. He also believes the switch will help patients better understand their results by reporting them in the units they're used to seeing every day. However, he added that the new initiative is not yet cast in stone. ADA is eagerly awaiting the final analysis of the ADAGE data in order to reach a final conclusion.

Change would require recalibration

Should the initial findings hold up, ADA, EASD, and the International Diabetes Foundation will then draft an outline for converting A1c results into ADAG units. If implemented, the change would require recalibration of machines, with subsequent simultaneous reporting of both the A1c and the new ADAG level by laboratories-a change that most likely would not occur until sometime in 2008. In addition, should the modification occur, ADA plans to launch an educational campaign to ensure that patients with diabetes learn about the change and the importance of the new value.

"Everyone is aiming for a chronic glucose level as close to the nondiabetic as possible," according to Nathan. He hopes the new measurement will enable diabetes to be managed more constructively.

Final results of the ADAGE trial were presented at the EASD annual meeting in Amsterdam just before press time. Results showed at least a 90% correlation between the new reference standard and the average blood glucose. Using regression analysis, the standard error from the mean was 0.83 mmol/mol, indicating a strong correlation. Nathan said the relationship appears to be the same for both men and women, and Type 1 and Type 2 diabetes.