ADA 2008: Separate drug classes prevent development of type 2 diabetes in susceptible patients

Pioglitazone prevented progression to type 2 diabetes in patients with impaired glucose tolerance (IGT), reported Ralph DeFronzo, MD. The effect was marked: 81% fewer patients assigned to pioglitazone compared with placebo converted from IGT to type 2 diabetes in a randomized, double-blind, placebo-controlled clinical study.

IGT is characterized by insulin resistance and impaired insulin secretion; a small proportion of persons with IGT already have diabetic retinopathy and peripheral neuropathy, said Dr. DeFronzo, professor of medicine and chief of the diabetes division, University of Texas Health Science Center, San Antonio.

"Five big studies have demonstrated that thiazolidinediones preserve beta cell function and cause a durable reduction in hemoglobin A1c in type 2 diabetes," he said.

The study reported here included 602 persons with IGT as determined by an oral glucose tolerance test (2-hour plasma glucose of 140 to 199 mg/dL) who were randomized to pioglitazone, initiated at 30 mg/day and titrated to 45 mg/day, or placebo. To be eligible for the study, subjects had to have fasting plasma glucose greater than 95 mg/dL and at least one other high-risk characteristic.

At study entry, all subjects had a reduction of insulin sensitivity and a decrease in the insulin secretion/insulin resistance index compared with 102 controls with normal glucose tolerance who were matched to the study participants by age, body mass index, and gender. The study participants returned every 3 months for measurement of fasting plasma glucose and hemoglobin A1c, and annually for oral glucose tolerance tests.

At the end of 4 years, 45 subjects (6.8% per year) randomized to placebo developed diabetes compared with only 10 (1.5% per year) randomized to pioglitazone (p < 0.00001). A return to normal glucose tolerance occurred in 42% of the pioglitazone group vs. 28% in the placebo group (p < 0.0005). Two-hour plasma glucose levels also decreased significantly in the pioglitazone group compared with the placebo (p < 0.005).

Both insulin sensitivity and the insulin secretion/insulin resistance index improved from baseline in the pioglitazone group (P <0.0005 for each) and were unchanged in the placebo group.

Sixteen bone fractures, eight in each group, occurred throughout the study, all related to trauma. Edema occurred in 22% of the group assigned to pioglitazone and 15% assigned to placebo, congestive heart failure developed in 0.3% of each group.

Based on these results, the number needed to treat to prevent one case of type 2 diabetes developing in patients with IGT was 3.5, "a very favorable number," said Dr. DeFronzo.

A novel investigational anti-inflammatory antioxidant compound called succinobucol also prevented progression from impaired or abnormal fasting glucose to type 2 diabetes, said Jean-Claude Tardif, MD, director of the Montreal Heart Institute Research Centre and chair in atherosclerosis at the Canadian Institutes of Health Research.

Development of new-onset diabetes was a secondary endpoint of the study, which enrolled 6144 patients who had an acute coronary syndrome within the previous year. Subjects were randomized to placebo or succinobucol on top of standard care. There was no significant difference in the two arms on the primary endpoint: a composite of coronary heart disease death, cardiac arrest, myocardial infarction (MI), stroke, the need for revascularization, or admission for unstable angina. However, those randomized to succinobucol had a significant 19% reduction (

p

= 0.029) on the composite of "hard" endpoints: cardiovascular death, cardiac arrest, MI, and stroke.

In a prespecified analysis among those without diabetes at the start of the trial, succinobucol was associated with a 63% reduction (p < 0.0001) in the development of new diabetes: 1.6% of the succinobucol recipients developed type 2 diabetes compared with placebo recipients.

In a post hoc analysis, subjects with impaired fasting glucose and abnormal fasting glucose at baseline were assessed for the development of diabetes. This analysis revealed a 60% reduction in the development of type 2 diabetes among these subjects randomized to succinobucol instead of placebo (p < 0.0001).

Unlike that observed with probucol, succinobucol had no effect on the QT interval or QTc, said Dr. Tardif. "This is clearly different than probucol," he said. However, those assigned to succinobucol were more likely to develop atrial fibrillation than the placebo group.

There were no increases in body weight or the incidence of edema or hypoglycemia with succinobucol compared with placebo.