Chicago, IL—At the 2013 American Diabetes Association (ADA) annual meeting, John E. Anderson, MD, President of Medicine & Science of the ADA, hosted the President’s Oral Research Sessions I and II, one session focused on basic science and the other on clinical and behavioral science. A subcommittee for the ADA’s Program Committee selected the abstracts of the 2 sessions. Commenting on these sessions, Dr Anderson said, “The Presidential Oral Research Sessions highlight the best and most promising research of all the abstracts submitted to the Scientific Sessions Planning Committee this year.”
A selection of the studies presented at these sessions are highlighted here.
Triple-Drug Combination as Initial Therapy Outperforms Sequential Therapy in Type 2 Diabetes
Beginning treatment with triple therapy compared with usual sequential therapy produced a faster and more sustained reduction in hemoglobin (Hb) A1c levels in patients with new-onset type 2 diabetes. The single-center, randomized study was presented by Muhammad A. Abdul-Ghani, MD, PhD, Assistant Professor, University of Texas Health Science Center, San Antonio.
Dr Abdul-Ghani presented the 2-year interim results from a 3-year study that seeks to determine the best way to attain the ADA/European Association for the Study of Diabetes–recommended goal of HbA1c <6.5% without hypoglycemic events, which has not been studied previously.
The initial triple therapy (N = 71) comprised metformin 1000 mg daily, pioglitazone 15 mg daily, and exenatide 5 μg twice daily (uptitrated at the end of month 1 as needed to 2000 mg, 45 mg, and 10 μg, respectively). Conventional therapy (N = 76) included metformin followed by glyburide in month 2 and insulin glargine in month 3. The initial starting daily doses of 1000 mg, 5 mg, and 10 units, respectively, were uptitrated if fasting plasma glucose (FPG) was >100 mg/dL or HbA1c >6.5%.
The patients (average age, 46 years) had an average body mass index of 36 kg/m2 and a 5-month average duration of diabetes since diagnosis.
With triple therapy, HbA1c was reduced to 6.0% at 6 months (from 8.6% at baseline) and remained at this level at 24 months. By contrast, although the HbA1c level was reduced to 6.1% with conventional therapy at 6 months, it rose to 6.6% at 24 months.
Postprandial hypoglycemia was markedly lower with triple combination therapy than with conventional sequential therapy and was the major contributor to achieving HbA1c goals. FPG was also significantly lower with triple therapy compared with conventional therapy.Other key improvements with triple combination therapy versus conventional therapy were:
- More patients achieved HbA1c <6.5% (60% vs 27%, respectively)
- More patients achieved HbA1c <7.0% (92% vs 72%, respectively)
- A lower rate of hypoglycemia (15% vs 46%, respectively)
- A better weight profile (–1.2 kg vs +4.1 kg, respectively).
Treatment failure was only 17% at 2 years with the triple therapy compared with 42% with conventional therapy.
Dr Abdul-Ghani said that their approach of correcting the known pathophysiologic defects of insulin resistance (with metformin and pioglitazone) and beta-cell dysfunction (with exenatide) at the time of diagnosis is more effective, safer, and more durable than targeting lowering plasma glucose concentration. “Treat the disease—diabetes—not its symptom of hyperglycemia. Treat the patient, not his blood sugar,” he said.
New Analysis of ACCORD: Insulin Exposure Not Linked to CV Mortality
The results of the ACCORD trial showed that intensive treatment of patients with type 2 diabetes who also had high cardiovascular (CV) risk resulted in increased CV and all-cause mortality. A new analysis of the ACCORD data was presented at the session by Elias S. Siraj, MD, Associate Professor of Medicine, Temple University School of Medicine, Philadelphia, PA. He and colleagues analyzed insulin exposure data from the ACCORD study for 10,163 patients with follow-up data (mean, 5 years) for CV mortality and insulin dose.
The unadjusted data showed an 83% higher risk of CV mortality for every 1-unit-per-kg increase in the insulin dose. However, no association was found between exposure to insulin and increased risk for CV mortality, after adjustment for baseline characteristics. The investigators concluded that after adjusting for 14 baseline characteristics and other covariates, including HbA1c and hypoglycemia, there was no support for the notion that insulin dose is an independent risk factor for CV mortality in this patient population.
Education, Not Technology, Reduces Hypoglycemia in Type 1 Diabetes
A structured 2-hour education program reduced the incidence of biochemical hypoglycemia in high-risk patients with type 1 diabetes, whether they were receiving insulin by a continuous subcutaneous insulin pump or by multiple daily injection. The Hypo COMPaSS trial is the first to investigate the management of impaired awareness of hypoglycemia, which is characteristic of approximately 20% of patients with type 1 diabetes and causes a 6-fold increase in severe hypoglycemia. Stuart A. Little, MBBS, Diabetes and Endocrinology Specialist Trainee, Newcastle University, England, presented the study results.
The patients were randomized to continuous subcutaneous insulin with insulin aspart (N = 46) or to multiple daily injection with insulin aspart plus insulin glargine (N = 50), and to continuous glucose monitoring or not.
Mr Little noted that all the patients (mean age, 49 years) were provided equal support to reach the treatment goal of avoiding biochemical hypoglycemia, but without relaxing the HbA1c goals. The patients’ mean diabetes duration was 29 years and the mean baseline HbA1c was 8.2%.
The primary end point of the change in the Gold score—an established measure of awareness of the onset of hypoglycemia—improved significantly (P <.001) in all participants, from a score of 5 to 4 (a lower number indicates greater awareness). There was no significant difference between groups.
By week 4, all patients achieved a significant reduction in biochemical hypoglycemia, from 53.3 minutes daily to 24.5 minutes daily, as was measured by continuous glucose monitoring; this reduction was sustained through week 24.
A reduction in the number of episodes of severe hypoglycemia was also achieved, from 9 at baseline to <1 at 24 weeks, and the proportion of patients affected was reduced to 19% compared with 92% at 1 year before the study.
Mr Little stated that the “benefit of education trumped the technology of insulin delivery” to avoid severe biochemical hypoglycemia.
Cognitive Decline in Type 1 Diabetes Quantified, Related to Systolic BP and HbA1c
The first study to assess changes over time in brain function and structure showed that there is an accelerated decline in executive functions and brain volume in patients with type 1 diabetes and microangiopathy. However, these cognitive changes are not comparable with the mild cognitive impairment known to lead to dementia. Eelco van Duinkerken, MSc, of the Diabetes Psychology Department, VU University Medical Center, Amsterdam, the Netherlands, presented the 4-year data at the President’s Oral Session.
In the 25 patients with type 1 diabetes and 25 healthy matched controls, baseline systolic blood pressure (BP) was related to a decline in brain volume, and baseline elevated HbA1c was related to a decline in executive function (ie, information processing, attention, memory). No correlation was found with severe hypoglycemic events.
The patients (mean age, 45 years) with diabetes had proliferative retinopathy as a marker of angiopathy, and their mean diabetes duration was 35 years. The systolic BP was 133.9 mm Hg in the diabetes group compared with 126.9 mm Hg in the control group (P = .045). The baseline HbA1c values were 7.9% and 5.4% (P <.001) in the diabetes and control groups, respectively.
As measured by magnetic resonance imaging, brain volume decreased by 1.34% in the patients with type 1 diabetes and by 0.68% in the control patients (P = .036). The decrease was most notable in the frontal and central areas of the right hemisphere, which are associated with executive function performance (P = .021).